<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Aeroz on Medium]]></title>
        <description><![CDATA[Stories by Aeroz on Medium]]></description>
        <link>https://medium.com/@aeroz.io?source=rss-e00a2bca2ed7------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*8aD9WAfeh4l0FDKNsI0CzA@2x.jpeg</url>
            <title>Stories by Aeroz on Medium</title>
            <link>https://medium.com/@aeroz.io?source=rss-e00a2bca2ed7------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 12:42:55 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@aeroz.io/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Forty-One Days Until Your Coffee Needs a Reference Number]]></title>
            <link>https://medium.com/@aeroz.io/forty-one-days-until-your-coffee-needs-a-reference-number-f8d595957bee?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/f8d595957bee</guid>
            <category><![CDATA[coffee]]></category>
            <category><![CDATA[dpp]]></category>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Thu, 19 Nov 2026 18:22:15 GMT</pubDate>
            <atom:updated>2026-11-19T18:22:15.433Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Cattle, cocoa, coffee, palm, rubber, soya and wood. Twice postponed, and the legislative file is now closed.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RSvxQPbxLPyW6RuYM-iWXQ.jpeg" /></figure><p>Aeroz <strong>Editorial</strong> November 2026</p><p>From 30 December, a customs declaration for cocoa, coffee, cattle products, palm oil, rubber, soya or wood that does not carry a valid due diligence statement reference number will not be accepted. That is forty-one days. However, the regulation has already been postponed twice.</p><p>Which is why the only question worth answering first is not what the EU Deforestation Regulation requires. It is whether you should believe the date.</p><h3>Do you believe it this time</h3><p>If you built for the original date and then for December 2025 and watched both evaporate, scepticism is the rational position and a third delay would cost you nothing to assume. So here is the procedural record rather than an opinion.</p><p>The postponement to 30 December 2026 was made by Regulation (EU) 2025/2650 of 19 December 2025, published in the Official Journal on 23 December 2025 and in force three days later. That is a completed act, not a proposal.</p><p>The European Parliament’s legislative train, which tracks the file as procedure 2025/0329(COD) and was last updated on 25 October 2026, records its status as “Adopted / Completed” and its most recent event as “Final act published in Official Journal.” It lists no further amendment, no pending postponement, no scheduled vote and no Council position.</p><p>The amending regulation also obliged the Commission to carry out a simplification review by 30 April 2026. That was the obvious vehicle for another delay. The Commission published it on 4 May 2026, five days late, and what came out was a report, updated guidance, an FAQ and a draft delegated act on product scope. Guidance, not postponement.</p><p>So the honest answer is that the mechanism that produced the last two delays has run its course and produced paperwork instead. Nothing in the legislative record as of late October suggests a third. That is not a guarantee, and the section at the end of this piece says what we could not check.</p><h3>Who the regulation thinks you are</h3><p>Before the data, settle the identity question, because a surprising number of companies have it wrong.</p><p>The obligation to file a due diligence statement falls on the operator: the party that first places the product on the EU market, or exports it. That is a legal role defined by the regulation, and it is not automatically the shipper, the consignee, the importer of record, or whoever arranges your customs clearance.</p><p>If you have assumed your freight forwarder or your customs broker carries this, they do not. If you have assumed your EU subsidiary does it because it receives the goods, check whether it is actually the entity placing them on the market.</p><p>The 2025 amendment sharpened this in a way that helps most readers. Only the operator who first places the product on the market files a full statement. Downstream operators and traders do not file one per product. The duty to collect and retain the upstream statement’s reference number falls on the first downstream operator, and Stibbe’s reading is that records of suppliers, customers and reference numbers are kept for five years.</p><p>So most companies in the chain inherit a number and a record-keeping duty. One company at the top does the work.</p><h3>What has to exist on 30 December</h3><p>For the operator filing, the three-step due diligence is unchanged by the amendment: gather the information, assess the risk, and mitigate until it is negligible.</p><p>The information includes the country of production and the geolocation coordinates of the plots of land where the commodity was produced. That is the part with no shortcut in it. A statement is filed in the Commission’s TRACES system, which reopened in June 2026 and now supports API submission and groups reference numbers under a single identifier, and a reference number comes back.</p><p>Two pieces of relief worth knowing. A statement can be filed annually per product rather than per consignment, though it has to be re-run when suppliers, geolocations or harvest dates change. And micro and small primary operators selling directly in low-risk countries make a one-off simplified declaration and may give postal addresses instead of coordinates.</p><p>Then the number has to reach the border. Maersk’s customer guide of 11 November states it plainly: customs declarations for goods in scope that do not contain a valid reference number will not be accepted, and the number should reach your broker before the shipment arrives rather than when it is already on a quay.</p><h3>What the amendment took off your plate</h3><p>Worth reading before you fund anything, because some of what people are still planning for is no longer required.</p><p>Micro and small enterprises, established as such by 31 December 2024, have until 30 June 2027, seven months after everyone else.</p><p>Printed products came out of Annex I entirely: books, newspapers, pictures and other printing-industry products.</p><p>Delegated Regulation (EU) 2026/2102 of September 2026 exempted waste, used and second-hand products, packaging, marketing and information materials, correspondence, products used in making medicinal products, and samples for testing. It also pushed a set of newly added products, including soluble coffee, certain palm oil derivatives and frozen cattle tongues, to 30 December 2027.</p><p>And the downstream relief above is the largest single reduction in the regime: most of the companies that spent 2024 preparing to file statements will not be filing them.</p><h3>What we could not settle</h3><p>Three things, and they are the ones a reader in luxury or leather will care about most.</p><p>The Commission’s own EUDR page would not load for us, so everything here is taken from the Parliament’s legislative train, the amending regulation as those sources describe it, a law firm note and a freight forwarder’s customer guide. The Council’s press release on the revision was also unreachable.</p><p>We could not confirm whether finished leather goods are in scope. Cattle is one of the seven commodities and the scope of derived products is set by Annex I, which we did not read. A handbag maker should not take this article’s word for it in either direction.</p><p>We could not establish the required precision of the geolocation data, nor whether coordinates are provided per plot or per consignment in every case, nor how the country benchmarking into low, standard and high risk changes the obligation in practice.</p><p>The penalty figure in circulation, up to 4 per cent of EU annual turnover alongside confiscation and exclusion from public procurement, comes from the freight forwarder’s guide and we did not confirm it against the regulation.</p><h3>What would have had to be provable, at item level</h3><p>The filing is not where this breaks. An operator with one supplier and one plot can do the whole thing in an afternoon.</p><p>It breaks where commodities mix. Coffee from forty smallholdings arrives as one lot. Rubber is blended. Hides from several abattoirs become one batch of leather. The regulation wants the statement tied to the plots of land where production happened, and the customs declaration wants a reference number tied to the consignment in front of it, and in between sits every blend, split, repack and re-lot your operation performs. The question the system has to answer is which plots are in this box, and that is lot genealogy: a link from input material forward into the finished units built from it.</p><p>We wrote earlier this month about a pharmaceutical firm that found metal contamination in a lot it never shipped and was able to recall exactly one released lot, because it could follow a quantity of intermediate material forward into the finished goods that carried it. That is the same capability. Here it is being asked for by a customs system instead of a quality director, and it has a date on it.</p><p>The firms that will file these statements cheaply are the ones that already know which inputs went into which outputs. The ones that will find December expensive are the ones that can reconstruct it, slowly, from paperwork. If you want to know which of those you are before the shipments start, the recall width assessment takes three minutes and collects nothing:</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f8d595957bee" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Ninety-Two Days Until Your Battery Needs a Registered Passport]]></title>
            <link>https://medium.com/@aeroz.io/ninety-two-days-until-your-battery-needs-a-registered-passport-514b7daaa126?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/514b7daaa126</guid>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Wed, 18 Nov 2026 23:45:57 GMT</pubDate>
            <atom:updated>2026-11-18T23:45:57.940Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Seventy-one data points, but carbon footprint and due diligence are not among them yet. The registry is already open.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fjoLnbKTVqf7fnwk4TxrPw.png" /></figure><p><a href="http://aeroz.io/">Aeroz</a> <strong>Editorial</strong> November 2026</p><p><strong>On 18 February 2027, an EV battery, a light-means-of-transport battery or an industrial battery over 2 kWh cannot be placed on the EU market without a passport registered in a Commission database.</strong> That is ninety-two days. The database has been open since July.</p><p>This piece is for the person who has to deliver that. It is not about anybody getting caught. It is a read of what the Commission has actually published, what it has not published yet, and which items in it have a lead time longer than the time remaining.</p><h3>What is required on the day</h3><p>The Commission’s guidance is called “Guidance to support preparations for the Digital Batteries Passport,” published by DG GROW on 21 August 2026, with the underlying document, “Digital Batteries Passport: data points by category,” now at version 3.2 and last updated on 19 October.</p><p>It sets out <strong>71 data points</strong> across EV, light-means-of-transport and industrial batteries. Each one is marked as mandatory, optional, conditional, or not required as of February 2027. That last category is the most useful column in the document and the one worth reading first.</p><p>One caution the guidance puts on itself, and it belongs in your project plan rather than a footnote: it “does not introduce additional legal requirements and should not be regarded as an authoritative interpretation.” It is the Commission helping you prepare. It is not the law, and the law is Regulation (EU) 2023/1542.</p><p>On scope, the Commission’s own pages describe it two ways, and they reconcile once you read the definitions. The news item says EV, LMT and industrial batteries over 2 kWh. The battery page says EV, e-bikes, e-mopeds, e-scooters, home storage and industrial. Those are the same set: an e-scooter battery is a light means of transport battery, and the Regulation treats batteries for energy storage in private or domestic environments as industrial.</p><h3>The three things not required in February</h3><p>This is the part most likely to change what you do this quarter, and it runs the other way from the usual deadline panic.</p><p>Per the Commission’s battery FAQ, <strong>carbon footprint, recycled content and due diligence fields are not required from February 2027.</strong> The due diligence report applies from August 2027, six months later.</p><p>Those three are the hardest data to assemble, because they reach back up a supply chain you do not control. If your programme has been sequenced around them, it has been sequenced around the wrong things, and there are ninety-two days in which that matters.</p><p>The implementing acts setting the carbon footprint declaration format are, in any case, still to come.</p><h3>Who has to do it</h3><p>The obligation sits with <strong>the economic operator placing the battery on the EU market or putting it into service</strong>. That operator registers the passport.</p><p>A non-EU operator can create an account, but for batteries the registration itself has to be done by whoever places the battery on the EU market. If you manufacture outside the EU and sell through an importer, the registration is the importer’s act and the importer’s liability, which is a conversation worth having before February rather than in March.</p><p>Article 77(4) of the Batteries Regulation allows a written authorisation for a third party to create or update the passport on your behalf. The legal responsibility does not move. It stays with the operator placing the battery on the market, whoever does the typing.</p><h3>The item with the longest lead time</h3><p>Here is the one that is not in most project plans.</p><p>You cannot register anything until your organisation is verified, and verification is not a form. An organisation verifies with a <strong>qualified electronic seal under eIDAS</strong>, or by evidencing identity and establishment through a qualified electronic attestation of attributes. A sole trader or natural person uses a qualified electronic signature or an eID at the ‘high’ assurance level.</p><p>A qualified electronic seal is issued by a qualified trust service provider, and obtaining one involves identity verification of the legal entity and its representatives. It is not something you complete on a Thursday afternoon.</p><p>Verification then lasts up to three years, so it is a one-off, but it is a one-off that sits in front of everything else. If nobody in your organisation has started it, that is the first item, not the data model.</p><p>Enrolment is open now at the Commission’s registry, and there is a separate test environment that takes its own EU login and touches no live data. The Commission runs a helpdesk at <a href="/service/mailto:EC-HELPDESK-DPP@ec.europa.eu">EC-HELPDESK-DPP@ec.europa.eu</a> and on +32 2 296 0431, between 08:00 and 20:00 CET. There is a fourth battery passport webinar on 26 November.</p><p>There is no deadline to pre-register. The obligation is that the passport is registered before the battery is placed on the market, which for a product you ship continuously means the capability has to exist before February rather than the entries.</p><h3>What is still missing, with ninety-two days to go</h3><p>An honest reading of the Commission’s own material shows three things that are not finished.</p><p><strong>The access-rights implementing act for batteries has not been adopted.</strong> This is the act that determines which data is public, which goes to persons with a legitimate interest such as repairers, remanufacturers and recyclers, and which is restricted to notified bodies, market surveillance authorities and the Commission. The FAQ indicated a draft was expected for feedback in October or November. We could not find a published draft. The Regulation’s recitals describe the architecture, including that test report results should reach only notified bodies, market surveillance authorities and the Commission, but the operative instrument is not law yet.</p><p><strong>Two of the eight harmonised standards are outstanding.</strong> Six have been published, and the Commission’s own timeline put an implementing decision on the remaining two in November 2026. We could not confirm it has happened.</p><p><strong>And there is an unresolved point about who mints the identifier.</strong> The general digital product passport page says the registry generates a unique URI. The battery FAQ says economic operators can create identifiers locally using options in the relevant standards, or use a commercial service. Those are different answers to an architectural question, and if you are building the integration you need the real one. It is a helpdesk question.</p><h3>A correction to our own reporting</h3><p>On 17 November we wrote that customs can verify electronically that an imported product has a valid registered passport before release for free circulation. That is the Commission’s general statement about digital product passports and we reported it as such.</p><p>For batteries specifically, the FAQ says something different: the passport’s <strong>unique registration identifier “shall not be entered into the electronic customs declaration.”</strong> The passport must still be registered. It is simply not a field in the declaration.</p><p>The FAQ also says dynamic fields can be empty at import and the passport can still be registered, which matters for anyone importing cells and completing data later.</p><p>We were not wrong about the general rule, and we should have checked the battery-specific FAQ before applying it to batteries. Corrected here.</p><h3>What it would have cost with unit-level identity</h3><p>The registry entry is not the hard part. Enrolment is a week of somebody’s attention and a trust service provider’s invoice.</p><p>The hard part is that this is a record <strong>per battery</strong>. Seventy-one data points attached to an individual unit, resolvable from a QR code on that unit, maintained after it leaves you. Some of it is static and some is not: the Regulation contemplates read-only access to battery management system data for state of health and expected lifetime, updated at least daily. That is not a document you produce once. It is a live record keyed to a serial number for the service life of the battery.</p><p>Most firms approaching February are trying to build that out of an ERP extract and a spreadsheet, because the deadline reads like a reporting obligation. It is not a reporting obligation. It is a per-unit identity with a data pipeline behind it, and the organisations that already assign and resolve a unique identity per unit are doing a data-mapping exercise, while the ones that do not are building the identity layer and the passport at the same time, in ninety-two days.</p><p>Which is the useful question to ask yourself this week: when a regulator, a recycler or a buyer scans one battery that left your plant eighteen months ago, what resolves, and who keeps it current? If you want to know what your records could answer today, the recall width assessment takes three minutes and collects nothing: <a href="https://claude.ai/artifact/BTXR16P7uGbA56i4t2PFMg">https://claude.ai/artifact/BTXR16P7uGbA56i4t2PFMg</a></p><p>The Regulation does not require NFC, and a QR code satisfies it. We also do not generate the battery data; your battery management system does.</p><h3>What we are not saying</h3><p>We are not saying the deadline will move. It is in a regulation in force, not a proposal, and nothing we read suggests a delay. But the ESPR textile and iron and steel timetables have already shifted and an access-rights act is still unadopted, so a reader planning capital spend should watch the Official Journal rather than take our word for it.</p><p>We are not stating the penalty for non-compliance, because the Commission’s pages do not. The Batteries Regulation leaves penalties to Member States, which are to make them effective, proportionate and dissuasive. The practical consequence on 18 February is about market access rather than a fine: a battery without a registered passport cannot lawfully be placed on the market.</p><p>And we could not read Article 77 itself. <strong>EUR-Lex served us the recitals and the first three articles of Regulation (EU) 2023/1542 and then truncated</strong>, across five attempts at different offsets. Everything here about the operative requirement comes from the Commission’s own guidance, battery page and FAQ, plus the Regulation’s recitals, which are interpretive rather than binding. Before you commit a budget, read Article 77 and Annex XIII in the Official Journal.</p><p>Aeroz tracks counterfeit enforcement, recalls and supply chain failures in regulated industries, three times a week. Follow for the next one. aeroz.io · <a href="/service/mailto:info@aeroz.io">info@aeroz.io</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=514b7daaa126" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From February Europe Can Refuse a Product That Cannot Identify Itself]]></title>
            <link>https://medium.com/@aeroz.io/from-february-europe-can-refuse-a-product-that-cannot-identify-itself-6354efe2df6c?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/6354efe2df6c</guid>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Tue, 17 Nov 2026 22:41:18 GMT</pubDate>
            <atom:updated>2026-11-17T22:41:18.766Z</atom:updated>
            <content:encoded><![CDATA[<p><em>The registry has been live since July. The first hard deadline is 18 February 2027, and the law behind it is late.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KMucSFV24BAJ0h5hz4jm0A.png" /></figure><p><a href="http://aeroz.io/">Aeroz</a> <strong>Editorial</strong> November 2026</p><p>From 18 February 2027, certain batteries cannot legally be placed on the European market unless a digital passport for them is registered in an EU database. It has been open since 20 July, and customs can already query it before goods are released.</p><p>The batteries are the first. Iron and steel are scheduled for a delegated act in the last quarter of this year, textiles, tyres and aluminium in 2027, furniture in 2028, mattresses and ICT products in 2029, each with at least eighteen months before the obligation bites.</p><p>That is the part everyone in a compliance function already knows, and it is not why this is worth your morning. The interesting question is the one the regime quietly depends on and has never answered: when a product arrives at the European border with no passport and no record, who is the authority supposed to call?</p><p>Europe’s own numbers say that for a large share of what arrives, the answer is nobody.</p><h3>What is actually live</h3><p>The European Commission published a press release on 20 July 2026 under the headline that the Digital Product Passport Registry is now live. Its text is unglamorous: economic operators “must register each Digital Product Passport in the Registry,” and registration “is available through either a secure user interface or an application programming interface (API).”</p><p>The registry does not hold your product data. The Commission describes it as storing “unique identifiers, registration data, and high-level metadata,” with the product information itself staying with the operator, who “may host these themselves or through DPP service providers.”</p><p>The legal basis is the Ecodesign Regulation, Regulation (EU) 2024/1781. The battery deadline comes from a different instrument, the Batteries Regulation, Regulation (EU) 2023/1542, and covers electric vehicle, light means of transport and certain industrial batteries.</p><p>One detail matters more than its length suggests. On the carrier, the Commission says only that “a digital data carrier, such as a QR code, links the physical product to its DPP.” It does not mandate a technology. A printed QR code satisfies it.</p><h3>The border, and the person standing at it</h3><p>Picture the officer, because this is where the regime stops being paperwork.</p><p>Under the customs arrangements the Commission describes, an officer can verify electronically that an imported product has a valid registered DPP and that the commodity code has been provided, before the goods are released for free circulation. That is a database lookup, not an inspection. It takes no laboratory, no sample and no expertise in batteries.</p><p>Which makes it the first enforcement tool in this field that actually scales to the volume. And the volume is the whole problem. Low-value consignments into the EU ran to 1.4 billion in 2022, 2.4 billion in 2023 and 4.6 billion in 2024, according to figures the European Parliament’s research service attributes to the single market strategy. For 2025 the Commission’s own customs reform page gives 5.9 billion, over ninety per cent of it from China. E-commerce now accounts for 97 per cent of all customs declarations.</p><p>Against that, physical checking has been going the other way. Customs checks fell to 82 per million imports in 2024, from 124 in 2023 and 203 in 2022. Market surveillance authorities checked 102 products per million inhabitants in 2024. The Commission’s own 2026 single market report calls the level of market surveillance insufficient.</p><p>So a lookup that can run on every consignment is worth more than an inspection regime that reaches almost nothing. That is the case for the passport, and it is a good one.</p><h3>What Europe found when it looked</h3><p>Three sets of figures, all from the European Parliament’s research service briefing on the European product act, each attributed there to a Commission source.</p><p>In a sweep announced on 18 July 2025, the Commission reported that around 52 per cent of checked listings carried compliant manufacturer information, rising to 65 per cent on very large online platforms. On information about the responsible person established in the EU, 68 per cent of listings were compliant, and 81 per cent on the very large platforms. After a product was removed by a very large platform, the briefing records, it reappeared in six cases.</p><p>In a priority control area running from April to June 2025, covering toys and small electronics bought from four major online marketplaces, 55 per cent of products were non-compliant overall and 84 per cent of those lab tested were found to be dangerous.</p><p>And Safety Gate, the EU’s alert system, logged 4,671 alerts for dangerous non-food products in 2025, up 13 per cent on 2024 and 121 per cent on 2022. Of those, 2,268 concerned products from Asian countries and 2,006 specifically from China.</p><p>Read the second figure again, because it is the one people skip. Not 84 per cent of a random sample. Eighty-four per cent of the products that were selected for laboratory testing turned out to be dangerous.</p><h3>The gap the passport sits on top of</h3><p>Here is the structural problem, and it is the clearest single fact I found this week.</p><p>BEUC, the European consumer organisation, published a checklist in June 2026 setting out what it wants from the forthcoming law. In it, BEUC states that the Market Surveillance Regulation requires an economic operator established in the EU for only 20 of 71 harmonised product categories. For non-harmonised products under the General Product Safety Regulation, the same requirement does not apply at all.</p><p>Twenty of seventy-one. For the rest, there is no legal obligation that anybody inside the European Union be answerable for the product.</p><p>A digital product passport is a record created and registered by an economic operator. If no economic operator is required to exist, the passport regime has a hole in the middle of it exactly where the parcels are arriving.</p><p>BEUC’s ask follows from that: online marketplaces should become economic operators where no manufacturer, importer or authorised representative in the EU exists or can be found.</p><h3>The law that was due in September</h3><p>Which brings us to the European Product Act, and to the reason this piece leads with a battery deadline rather than a bill.</p><p>The Commission’s 2026 work programme set adoption of the package for the third quarter of this year. The Parliament’s research service noted in June that the likely date was September. Law firms tracking it expected publication on 6 October. Cooley’s own tracker now carries a banner saying publication has been postponed until later this year, date to be confirmed. As of today it has not appeared.</p><p>What is known comes from a leaked draft, which is a weaker thing than a bill and should be read that way. Three firms have published analyses of it, and they agree on the shape. The Act would merge the New Legislative Framework and the Market Surveillance Regulation, Regulation (EU) 2019/1020, into one directly applicable regulation. Noerr puts the draft at over 220 pages.</p><p>The provision that matters for everything above: a marketplace would have to verify, before a listing goes live, that the product has a valid digital product passport or product responsibility record and that the listing matches it. If it does not, and the product reaches the market with no EU-established operator responsible for it, the marketplace itself assumes the obligations of an authorised representative.</p><p>On penalties, the sources differ and the difference is worth noting rather than averaging. Cooley reads the draft as requiring member state penalties of between 1 and 6 per cent of global annual turnover, with separate Union-level penalties of up to 6 per cent. Noerr reports fines of up to 6 per cent of global annual turnover. Nobody is quoting a published text, because there isn’t one.</p><p>At a European Parliament committee hearing in early October, a Commission official said the Commission is considering requiring every imported product to have a responsible operator within the EU. That is the gap being closed, said out loud.</p><h3>What it would have cost with unit-level identity</h3><p>Here is the honest commercial read, and it has two halves that point in different directions.</p><p>The first half is that this is a dated project with a known scope, and most firms in scope are treating it as a labelling exercise. The work is: assign a unique identifier to every unit, hold a record against it, publish a resolvable link, register the identifier. A company that does the minimum produces a QR code that opens a PDF. A company that does the same work once, properly, ends up with the thing that sets its recall width, because an identifier that resolves to a unit is the same asset whether a customs officer queries it or a quality director does at two in the morning. The deadlines are already written down. February for batteries, then a category a year.</p><p>The second half is the part our own industry keeps quiet about. The regulation does not require NFC, and it does not require anything we sell. The Commission’s wording is “a digital data carrier, such as a QR code.” A label printer satisfies the law. Anyone telling you the EU is mandating secure authentication is selling you something, and we are not going to be that.</p><p>What the passport establishes is that a registered record exists and that this code points at it. It does not establish that this physical item is the one the record describes. A QR code is a number printed on a label, and a number printed on a label can be printed again on another label. We covered the limits of what a scan proves earlier in this series and the regulation does not change them.</p><p>So the useful question for a compliance budget is not whether to comply. It is whether the identifier you create to satisfy February is one you can also rely on in a recall, in a customs dispute, or when somebody shows you a unit and asks whether it is yours. If you want to know which of those your current records support, the recall width assessment takes about three minutes and collects nothing.</p><h3>What we are not saying</h3><p>We are not saying the European Product Act will pass as drafted, or at all. It is an unpublished leaked draft, the proposal is already late, and five member states, the Netherlands, Finland, France, Portugal and Sweden, put a joint note to the Competitiveness Council on 22 May 2026 asking among other things that physical CE marking be kept, which is not the direction the draft goes. Legislative drafts change, and this one has not even been tabled.</p><p>We are not saying a digital product passport reduces counterfeiting. It makes the absence of a record detectable at a border, cheaply and at volume, which is a different and narrower thing.</p><p>And we are not saying the figures above are a measurement of counterfeit goods. They measure non-compliance and danger, which overlap with counterfeiting without being it. A product can be perfectly genuine, made by exactly who it says, and still fail a toy safety standard.</p><p><a href="http://aeroz.io/">Aeroz</a> tracks counterfeit enforcement, recalls and supply chain failures in regulated industries, three times a week. Follow for the next one. aeroz.io · <a href="/service/mailto:info@aeroz.io">info@aeroz.io</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6354efe2df6c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Customs Is About to Email You a Photo and Ask If It’s Yours]]></title>
            <link>https://medium.com/@aeroz.io/customs-is-about-to-email-you-a-photo-and-ask-if-its-yours-b75791c6a56a?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/b75791c6a56a</guid>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Sat, 14 Nov 2026 00:46:44 GMT</pubDate>
            <atom:updated>2026-11-14T00:50:14.890Z</atom:updated>
            <content:encoded><![CDATA[<p><em>H.R. 4930 adds the container and the packing materials to what CBP can share. EU brand alerts fell to 1,305.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YgJvIfx4a98RkJ7r5W92yw.png" /></figure><p><a href="http://aeroz.io/">Aeroz</a> <strong>Editorial</strong> November 2026</p><p>The House passed H.R. 4930 on 27 April 2026, a bill whose long title is to expand the sharing of information on suspected intellectual property violations in trade. It adds the shipping container and the packing materials to what Customs can photograph and send you.</p><p>Trade press reported on 13 November that the Senate passed it without opposition three days earlier, and that it now awaits the President’s signature. The industry associations welcomed it, correctly. It is a sensible bill and it solves a real bottleneck.</p><p>It also quietly restates something about how counterfeit enforcement in the United States actually works, and the restatement is the reason to read it.</p><p>Customs is not getting better at deciding what is fake. It is getting a faster way to ask you.</p><h3>What the bill actually changes</h3><p>The bill amends section 628A of the Tariff Act of 1930, at 19 U.S.C. 1628a. It is short. Three changes do the work.</p><p>First, it widens the list of things Customs can share. The existing law already refers to “images of the merchandise.” The bill strikes “and its packaging and labels” and inserts “, its packaging, packing materials, containers, and labels,” and separately inserts “, packing materials, containers,” after “its packaging.”</p><p>So the photograph is not new. What is new is the container and the packing materials, which is to say the parts of a shipment that tell you about the route rather than the product.</p><p>Second, it adds a new paragraph allowing Customs to provide “nonpublic information about the merchandise that was generated by an online marketplace or other similar market platform, express consignment operator, freight forwarder, or any other entity that plays a role in the sale or importation of merchandise.” That is the genuinely significant provision, and almost nobody is reporting it. Marketplace and carrier data, generated by commercial parties, passed to a private rights holder by the government.</p><p>Third, it widens who can receive all this, adding “any other party with an interest in the merchandise, as determined appropriate by the Commissioner.”</p><p>And one more line, easy to skip. In the matter preceding the list, the bill strikes “suspects” and inserts “has a reasonable suspicion.”</p><h3>The threshold change is not what the coverage says</h3><p>That last amendment has been described in trade coverage as lowering the bar. Read as text, it is not obviously either.</p><p>“Suspects” is a verb with no defined content. “Reasonable suspicion” is a term of art with a body of law behind it. Substituting the second for the first replaces an unbounded discretion with a standard a court could review. Whether that makes sharing easier or harder in practice depends on how Customs currently reads “suspects,” which is not something the amendment tells you.</p><p>We are flagging it because every secondary account we read treated it as a loosening, and the text does not say that. It is the kind of detail that gets repeated into a deck and then into a compliance assumption.</p><h3>What the bill does not do</h3><p>Here is the shape of it, and it is worth saying plainly.</p><p>There is no penalty in this bill. There is no duty on anyone who receives the information. There is no effective date provision. The only obligation it creates runs the other way: Customs must notify the recipient of the nonpublic information it transmits, under regulations the Secretary will write.</p><p>It is a pipe. It moves more information, from more sources, to more parties, faster.</p><p>Which leaves the question the pipe delivers to your desk.</p><h3>The photograph arrives. Then what?</h3><p>A container is held. An officer forms a reasonable suspicion. Under the amended section, you get images of the merchandise, its packaging, its packing materials, the container, and the labels, plus whatever the marketplace and the freight forwarder generated about the shipment.</p><p>Now somebody at your company has to look at a photograph of a box and say whether it is yours.</p><p>That is the whole system, and it has always been the whole system. This series has recorded the same structure in four jurisdictions. South Africa’s Counterfeit Goods Act has the brand supply the determination and the inspector test only whether the brand’s suspicion is reasonable. Ghana’s CID Director-General said in Accra last week that investigations “depend on forensic expertise, intelligence and cooperation with rights holders.” A Karnataka regulator holding a suspected counterfeit vial reportedly could not test it, because the analysis needed a reference standard only the brand owner could supply. And at Chicago O’Hare, when officers pulled two hundred Cartier watches, the parcel had to go to Customs’ own Centres of Excellence and Expertise before anyone could call them inauthentic.</p><p>The state has the goods. The state does not have the answer. The answer is held by a brand, in a building, by a person who may or may not be able to produce it quickly.</p><p>H.R. 4930 makes that request arrive faster and with better attachments. It does nothing about the reply.</p><h3>The number that should worry the people cheering</h3><p>Set the bill against the only measurement we have of how brands currently respond.</p><p>In the European Union, detentions rose sharply in the most recent reporting year. Over the same period, rights-holder alerts filed through the EU’s IP Enforcement Portal fell to 1,305 from 1,981, a drop of about a third. That figure comes from trade coverage of the EUIPO and Commission report rather than from the report itself, which we have not read, and we flag it as such because it is load-bearing here.</p><p>If it is right, it describes an uncomfortable direction. Enforcement capacity is going up. Brand participation is going down.</p><p>There are decent reasons for that. Alert programmes are administrative work, the volumes are enormous, and a brand protection team of four people cannot triage photographs all day. But the direction matters, because every statute in this field, including the one Congress just passed, assumes the rights holder is a willing and capable responder.</p><p>And the baseline is already thin. On the EU’s own figures, border detentions amount to 0.66 per mille of third-country imports. The system catches well under one per cent, and then asks the brand to adjudicate what it caught.</p><h3>The test, and it takes an afternoon</h3><p>Here is something concrete, and it costs nothing but a meeting.</p><p>Take one of your own products. Photograph the outer carton, the packing materials and the shipping container the way a Customs officer would: on a dock, under bad light, with no serial visible and no documentation attached. Send those images to whoever in your organisation would receive a Customs notification.</p><p>Then time how long it takes to come back with a defensible answer, and watch what they ask for.</p><p>If they ask for the lot number, you have learned that your identification capability depends on data that will not be in the photograph. If they ask to see the product itself, you have learned that your answer requires physical access that a notification does not give you. If they say yes or no in ten minutes with confidence, ask them what they were actually looking at, and whether a competent counterfeiter could reproduce it.</p><p>Most teams discover that what they identify is the artwork. Artwork is the thing counterfeiters have been getting right for a decade.</p><h3>This is the same question as the recall question</h3><p>We published a self-assessment this week built on a different number, recall width: how much product you would have to pull if a withdrawal were forced on you. It asks ten questions, each drawn from a documented case, and works out whether your answer would come down to units, lots, a production window, or everything still in date.</p><p>The photograph question belongs to the same family. Both are cases where somebody outside your company asks you something under time pressure, and your answer is determined entirely by records and capabilities you built, or did not build, long before.</p><p><a href="https://claude.ai/artifact/BTXR16P7uGbA56i4t2PFMg">How wide would your recall be?</a> Three minutes, nothing collected, no sign-up.</p><h3>What we are not saying</h3><p>We are not saying the bill is bad. It is a narrow, sensible amendment that removes a real friction, it passed both chambers without opposition, and the industry support for it is well founded.</p><p>We are not saying Customs should be adjudicating authenticity itself. It should not, it cannot at volume, and a system that routes the question to the party who actually knows the product is the right design. Our point is about the reply, not the request.</p><p>We are not saying brands are failing. The alert figure is one number, from a secondary reading of a report we have not opened, and declining participation has explanations that are not negligence.</p><p>And we are not saying our own field answers the photograph. It does not. A verification layer answers a question about a unit somebody is holding. Nobody at Customs is tapping your packaging during a notification, and a photograph of a carton is not a unit in anyone’s hand. What identity changes is upstream of this: fewer units that can present as yours in the first place.</p><h3>Who meets this problem</h3><p>The brand protection lead who will receive more notifications after this bill than before, with the same headcount.</p><p>The quality manager asked to confirm from an image, who knows that the only thing distinguishable in it is the print.</p><p>The customs officer who formed a reasonable suspicion, did everything right, and is now waiting on an email.</p><p>The marketplace or freight forwarder whose own nonpublic data is about to be handed to a rights holder, which is a provision nobody has written about and somebody should.</p><p>And the general counsel who will be asked, at some point, why a shipment was released.</p><h3>What would have had to be provable</h3><p>For the photograph to be answerable, the thing in it would have to carry something a camera can capture and a counterfeiter cannot reproduce. Artwork is not that. A printed code is not that, because a printed code photographs identically whether it is valid or copied.</p><p>What is actually provable from a picture is narrow, and being honest about that is more useful than pretending otherwise: you can establish that a carton is inconsistent with your current artwork, or that a container route is inconsistent with your distribution. Both are useful. Neither is proof.</p><p>The bill moves the question to you faster. The question was always going to arrive. The only variable that was ever in your control is whether anybody at your company can answer it.</p><p><a href="http://aeroz.io/">Aeroz</a> tracks counterfeit enforcement, recalls and supply chain failures in regulated industries, three times a week. Follow for the next one. aeroz.io · <a href="/service/mailto:info@aeroz.io">info@aeroz.io</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b75791c6a56a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DSCSA Measures You on 27 November. Are you ready?]]></title>
            <link>https://medium.com/@aeroz.io/dscsa-measures-you-on-27-november-are-you-ready-b0d0c4587b4a?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/b0d0c4587b4a</guid>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Fri, 13 Nov 2026 02:11:27 GMT</pubDate>
            <atom:updated>2026-11-13T02:11:27.629Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Large dispensers have been on package-level tracing for a year. The rule asks for data exchange, not a recall.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ICVs_chTtxGmvKZ9YS-rXw.jpeg" /></figure><p><a href="http://aeroz.io/">Aeroz</a> <strong>Editorial</strong> November 2026</p><p>On 27 November 2026, FDA measures which American pharmacies count as small dispensers: twenty-five or fewer full-time licensed pharmacists or qualified technicians at the owning company. Larger dispensers have been under full package-level tracing since the same date last year.</p><p>So the drug supply chain is now a year into the most detailed traceability regime the United States has ever required of it. Serialized package identifiers, electronic transaction information, interoperable exchange between trading partners, most of it running on EPCIS.</p><p>Here is the question nobody is being asked on 27 November, and it is the only one that will matter on the day something goes wrong.</p><p>Can you use any of it to run a recall?</p><h3>What the rule actually requires</h3><p>Be precise, because this is where the confidence comes from.</p><p>DSCSA requires trading partners to exchange transaction information and statements electronically, to verify product at the package level, and to operate systems that interoperate with everyone else’s. It is a data obligation, and it is a real one. A year in, large dispensers are exchanging that data daily.</p><p>What the rule does not require is a demonstration. Nowhere does it ask a company to prove that, given a lot number at nine in the morning, it can produce by nine the next morning a list of every customer who received a unit from it.</p><p>Those are different capabilities. The first is a format. The second is a query, run under time pressure, across systems that were built to satisfy the first.</p><h3>The gap is not theoretical</h3><p>Two things from the last few weeks make the point better than an argument would.</p><p>When FDA wrote to a contract manufacturer in Guangzhou on 27 October, one of its demands was a list of all products distributed in the United States in the last three years, with lot number, quantity shipped and expiry dates, plus the customers and addresses they went to. That is the query. The agency had to ask for it, in a letter, with a fifteen business day deadline, because it is not something a regulator can pull.</p><p>And when a toothpaste manufacturer in Haridwar recalled on 10 September, it recalled everything still within expiry that it had shipped across two years, because the records that would have narrowed it were the records under question.</p><p>Neither of those firms is a DSCSA dispenser. The point is not the jurisdiction. It is that in both cases the system contained a great deal of data and still could not answer the one question a withdrawal turns on.</p><h3>Why serialized data does not automatically answer it</h3><p>A serialized identifier tells you that a specific package exists and that a specific transaction happened. A recall question runs the other way: start from a lot, fan out to every package in it, then to every party that took custody, then to what each of them did next.</p><p>That reverse traversal is not what transaction exchange was designed for. It usually crosses several trading partners’ systems, each holding its own slice, in formats that satisfy a standard without being joined up for interrogation. Each party can answer for its own hop. Almost nobody has tested the whole chain end to end against a clock.</p><p>The honest version is this. DSCSA has made the data exist. It has not made the answer fast.</p><h3>Two pharmacies on the same street, two different answers</h3><p>There is a second consequence of the 27 November measurement that nobody frames as a recall problem, and it should be.</p><p>A dispenser owned by a company with twenty-six or more full-time licensed pharmacists or qualified technicians crossed into full package-level tracing on 27 November 2025. One at twenty-five or fewer is exempt from five of those enhanced requirements until 27 November 2027, and FDA urges them to prepare in the meantime but asks them to file nothing.</p><p>So from this month there are two populations of American dispenser, side by side, holding the same products. One can in principle answer at the package level. The other has no obligation to, for another year.</p><p>Now run a manufacturer’s recall through that.</p><p>If you are withdrawing a lot, you have to reach every party that received a unit, and a meaningful share of them are small dispensers with no package-level requirement. Your ability to narrow the recall is capped by the least capable party in the chain, not by your own systems. You can serialize perfectly, exchange flawlessly, and still end up asking a thousand independent pharmacies to go and look on a shelf.</p><p>That is not a criticism of the exemption. Small pharmacies had a real case for more time and FDA gave it to them for sound reasons. It is a statement about where your recall width actually lives, which is downstream of you, in other people’s systems, and largely outside your control.</p><p>Worth knowing before you describe your traceability programme to a board.</p><h3>Where the chain usually breaks</h3><p>When the query does fail, it fails in predictable places.</p><p>It fails at the hop into a third-party logistics provider, where the data is held under a different schema and reconciliation is a manual job. It fails where a wholesaler aggregates and the lot-level link is preserved for compliance but not indexed for search. It fails at returns and resupply, where a unit comes back into the chain and rejoins it with a new transaction history. And it fails at the last hop, where the receiving party records the shipment rather than the packages.</p><p>None of those is a scandal. Every one is a system doing exactly what it was specified to do. The specification was exchange, and the need is interrogation.</p><h3>The test to run before anybody makes you</h3><p>This one costs nothing and takes a morning.</p><p>Pick a lot number at random from six months ago. Not a convenient one. Ask your team for a list of every customer who received a unit from it, and time how long it takes to arrive complete.</p><p>If the answer comes back in an hour, you are in better shape than most and you should say so in your next audit. If it takes three days and a spreadsheet merge, that is your recall timeline, and you have just discovered it for free rather than during a withdrawal.</p><p>If the answer is partial, note exactly where it broke. That hop is the whole problem, and it is usually one trading partner or one system that was never joined to the others.</p><h3>What a good answer looks like</h3><p>For contrast, because this piece should not only describe a hole.</p><p>The companies that answer the query fast have three things, and none of them is exotic. They hold their own lot-to-customer mapping rather than relying on being able to reassemble it from trading partner exchanges. They have run the query at least once as a drill, with a stopwatch, on a lot nobody chose for them. And they know in advance which hop in their chain is the slow one, so the recall plan routes around it instead of discovering it.</p><p>That is a week of work and an owner, not a platform purchase. It is also the difference between a withdrawal that names lots and a withdrawal that names a year.</p><h3>This is one of ten</h3><p>That exercise is question four of ten in a short self-assessment we published this week. It works out your <strong>recall width</strong>: whether a withdrawal would come down to individual units, to lots, to a production window, or to everything still in date.</p><p>It takes three minutes. Nothing is collected, there is no sign-up and no form, and your answers stay in your browser.</p><p><a href="https://claude.ai/artifact/BTXR16P7uGbA56i4t2PFMg"><strong>How wide would your recall be?</strong></a></p><p>The reason it exists is that recall width is a cost figure, and it is fixed long before the recall. On the day a regulator calls, you cannot improve it. The questions are the decisions that set it.</p><h3>What we are not saying</h3><p>We are not saying DSCSA was a mistake or that the data is worthless. Serialization is the precondition for everything else, and a supply chain that can identify packages is better than one that cannot.</p><p>We are not saying anybody is failing. There is no published evidence that dispensers cannot answer the query, because as far as we can establish nobody has measured it. That absence is the point rather than an accusation.</p><p>And we are not saying the thing we build closes this. Unit-level identity tells you what a unit is when you are holding it. It does not assemble a distribution list across four companies’ systems on a Tuesday morning. What narrows a recall is a chain that can be traversed backwards, and that is an integration problem before it is a hardware one.</p><h3>One sentence to take into 27 November</h3><p>The rule measures headcount that day. The thing that will be measured on a worse day is how fast you can answer a question the rule never asked you to practice.</p><p><a href="http://aeroz.io/">Aeroz</a> tracks counterfeit enforcement, recalls and supply chain failures in regulated industries, three times a week. Follow for the next one. aeroz.io · <a href="/service/mailto:info@aeroz.io">info@aeroz.io</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b0d0c4587b4a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How Wide Is Your Recall?]]></title>
            <link>https://medium.com/@aeroz.io/how-wide-is-your-recall-ade2b5108106?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/ade2b5108106</guid>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Wed, 11 Nov 2026 22:54:55 GMT</pubDate>
            <atom:updated>2026-11-11T22:54:55.523Z</atom:updated>
            <content:encoded><![CDATA[<p>In the off-chance you need to recall product, would you even know where to start?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K3K2axNdn6H39xrt6FzMxQ.jpeg" /></figure><p><a href="http://aeroz.io/">Aeroz</a> <strong>Editorial</strong> November 2026</p><p>In September 2026 Ford recalled 11,405 Rangers over potentially counterfeit capacitors, and could name every truck. In the same month a contract manufacturer in Haridwar recalled every tube of toothpaste still in date it had shipped in two years, because it could not name one.</p><p>Neither company had a worse problem than the other. One of them had better records.</p><p>That difference has a name nobody uses, and it is the number that decides what a withdrawal costs: <strong>recall width</strong>. Not whether you have a problem. How much product you have to pull when you do.</p><h3>The two cases, side by side</h3><p>Ford’s recall is the one worth studying, because it is the only genuinely good outcome in a year of them.</p><p>A multi-layer ceramic capacitor on a centre-stack display board, from a tier-two supplier, failed both engineering and material specifications. A cracked capacitor shorts, blows the display fuse, and takes out the screen and the rearview camera feed with it. That is a safety defect, and the parts were described in the filing as potentially counterfeit.</p><p>What bounded it was a chain: the suspect capacitor lot connected to the printed circuit board’s serial number, the board to the centre stack assembly serial number, and the assembly to a vehicle. Lot to board, board to assembly, assembly to truck. The result was a recall with a build window, a truck count and a mailing list, and an estimate that about 47 per cent of those trucks actually carried the defect.</p><p>Now the other one. FDA’s warning letter of 27 October 2026 to Pritam International Private Limited, a contract manufacturer of over-the-counter toothpaste for the American market, records that investigators saw visible microbial growth on media plates in an incubator, and that the following day “there was no visible microbiological growth on plates labeled as the same samples observed the day before.” The firm’s own response described “a deliberate but isolated act by the microbiologist to circumvent your procedure.”</p><p>The letter also records that across its shipments of sorbitol, a material carrying a diethylene glycol contamination risk, the firm had one certificate of analysis showing the required identity results. FDA’s position is flat: “Identity testing, including DEG and EG, is required for each lot that you receive.”</p><p>On 10 September 2026 the company recalled all toothpaste still within expiry that it had shipped between September 2024 and August 2026.</p><p>Two years. Not because two years of product was bad, but because when the laboratory records cannot be relied on and incoming materials were not identity tested, there is no narrower answer available. A recall is only ever as precise as the evidence behind it, and the evidence was the problem.</p><h3>Width is set years before the recall</h3><p>Here is the part that matters to anyone reading this from inside a manufacturer.</p><p>On the day a regulator calls, your recall width is already fixed. You cannot improve it in the meeting. It was set by decisions taken quarters or years earlier, by people optimising for something else: whether to accept a supplier certificate instead of testing, whether to carry component serials through assembly, whether to hold a contract manufacturer’s batch records or just their certificates.</p><p>None of those decisions feels like a recall decision at the time. Every one of them is.</p><p>And the cost difference is not marginal. Ford pulled a defined population with an estimated defect rate under half. Pritam pulled everything it had shipped in two years, almost all of which was presumably fine. Same category of failure, radically different bill, and the variable was documentation.</p><h3>The four widths</h3><p>In practice there are four answers to “what would you have to pull”, and every manufacturer sits in one of them.</p><p><strong>Unit level.</strong> You can identify individual affected units. Rare, and it is the only position where the recall matches the fault.</p><p><strong>Lot level.</strong> You can bound it to batches. This is where most regulated manufacturers are, and it is a working position. The exposure is that a lot is a lot of units.</p><p><strong>Window level.</strong> You would pull a production window, because nothing in your records separates good from suspect inside it. Expect months.</p><p><strong>Everything in date.</strong> No narrower answer exists. If a supplier’s records are discredited, you withdraw all of it. This is where Pritam ended up, and this is where the eight-figure numbers live.</p><p>Most companies have never worked out which band they are in, which is odd given that it is a cost figure and finance would act on it immediately if anyone produced it.</p><h3>So we built the ten questions</h3><p>This is the part where an article usually tells you to think carefully about traceability and then ends.</p><p>Instead: <a href="https://aeroz.io/audits#audit"><strong>we built the assessment, and it is free and takes three minutes.</strong></a> Ten questions, each one drawn from a documented case in the last three years, answered about your own operation. It computes which of the four bands you are in and shows you the three answers that put you there.</p><p>Nothing is collected. No email, no sign-up, no form. The answers stay in your browser and the result renders as text you can copy into a meeting. We built it that way deliberately, because a lead form in front of a diagnostic is a tax on the people most likely to need it.</p><p>A few of the questions, so you know the shape:</p><ul><li>For your highest-risk incoming material, how many of your last twelve lots did you identity-test yourself, rather than accept on a supplier’s certificate?</li><li>Can you link a finished unit back to the specific lot of every critical component in it, without a manual file search?</li><li>If a regulator published an alert naming your product tomorrow, could you state what is actually in the units within 48 hours?</li><li>Have you confirmed in writing that your insurance covers a withdrawal compelled by a regulator?</li></ul><p>That last one catches people. A US court held in 2023 that a recall forced by the FDA was not an “accident” within the meaning of a general liability policy, so the policy was never triggered. The claimant had a $13 million consent judgment and recovered nothing on the insurance. If you have assumed your general liability covers a compelled withdrawal, that assumption is the single most expensive unchecked item on this list.</p><h3>What the assessment will probably tell you</h3><p>We have run the ten questions against every case in our own files from the last three years, and the pattern is consistent enough to predict.</p><p>Companies do well on customer traceability, because distribution is a commercial system and somebody owns it. They do badly on incoming material, because testing costs money and a supplier certificate is free. They do worst on the question of who physically made each component, as opposed to who invoiced them, which is the question contract manufacturing makes hard and nobody asks until a regulator does.</p><p>If you score badly on incoming material and well on everything else, you are in the window band and you probably think you are in the lot band. That gap is the one worth closing, and it is cheaper to close than most people expect, because it is usually three or four materials that carry the real risk rather than the whole bill of materials.</p><h3>What we are not saying</h3><p>We are not saying good records prevent failures. Ford’s capacitors still reached an assembly line. Records decide the size of the response, not whether there is one.</p><p>We are not saying Pritam is unusual. It is in the public record because FDA published a letter, which most comparable situations never produce.</p><p>We are not saying the assessment is a product pitch with a quiz on top. It gives you a result and a list of your own weakest answers. If it tells you that you are in good shape, that is the result, and you owe us nothing.</p><p>And we are not saying that what we sell closes all four bands. It does not. Unit-level identity narrows the width of what you can find and name in the field. It does nothing about whether the sorbitol was tested, and we have said so in this series repeatedly, including about cases that ended in criminal convictions.</p><h3>Who this is for</h3><p>The quality lead who could answer eight of the ten questions today and would rather know which two before somebody else asks.</p><p>The supply chain director who has a traceability project competing with four other projects and needs a number to defend it.</p><p>The CFO who has never seen recall exposure expressed as a population rather than a provision.</p><p>And anyone who read the Ford recall this year and assumed that level of precision was normal. It is not. It is the exception, and it was bought years in advance.</p><h3>Three minutes</h3><p>The assessment is here, it is free, and nothing is collected:<a href="https://aeroz.io/audits#audit"> <strong>How wide would your recall be?</strong></a></p><p>If the result is uncomfortable, the three weakest answers it gives you are the shortest list of things worth fixing that we know how to produce. If you want a second pair of eyes on those three, that is what we do.</p><p><a href="http://aeroz.io/">Aeroz</a>, Making Authenticity Undeniable. Visit <a href="https://aeroz.io/technology">aeroz.io</a> to learn more &amp; get in contact with our team via info@aeroz.io.</p><p>2026 AEROZ all rights reserved.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ade2b5108106" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Stimwave Made Its Own Fake Part. Every Check Said Genuine.]]></title>
            <link>https://medium.com/@aeroz.io/stimwave-made-its-own-fake-part-every-check-said-genuine-a1fae11125b6?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/a1fae11125b6</guid>
            <category><![CDATA[stimwave]]></category>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Mon, 09 Nov 2026 20:37:09 GMT</pubDate>
            <atom:updated>2026-11-09T20:37:09.534Z</atom:updated>
            <content:encoded><![CDATA[<p><em>PIP, Takata and Ranbaxy were the same failure. 67 million airbags, and none of it counts as counterfeiting.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n-xVnutbFVunlblxOs0XAA.jpeg" /></figure><p><a href="http://aeroz.io/">Aeroz</a> <strong>Editorial</strong> November 2026</p><p>Stimwave sold doctors a neurostimulator component called the White Stylet. It was plastic, it contained no copper, it could not receive anything, and on June 17, 2024 its creator was sentenced to six years for it.</p><p>This series has written thirty-nine pieces on the assumption that the counterfeiter is an outsider. Somebody who is not the manufacturer makes a thing that pretends to be the manufacturer’s. Every mechanism we have examined, printed codes, serialized barcodes, cryptographic tags, state verification marks, is built to answer that one question: did this come from who it says it came from.</p><p>The White Stylet did. That is the whole problem with it.</p><h3>What the device was</h3><p>The StimQ PNS System is an implantable neurostimulator for intractable chronic pain, producing electrical current to target peripheral nerves outside the spinal cord. It had three parts: an implantable electrode array called the Lead, an externally worn battery, and a receiver.</p><p>The receiver was the Pink Stylet. In the words of the sentencing release from the U.S. Attorney for the Southern District of New York on June 17, 2024, it “acted as an antenna, transmitting energy from an external power source to the Lead.” It contained copper. It was about 23 centimetres long, and that length was a problem: physicians told Stimwave they were having trouble implanting it in certain patients, and it could not be cut shorter without affecting its function as a receiver.</p><p>So a genuine engineering constraint produced a genuine commercial problem. A device that does not fit a patient does not get implanted, and a device that does not get implanted does not get billed.</p><h3>The part that was built to be billed</h3><p>What Stimwave produced in answer was the White Stylet.</p><p>The sentencing release describes it as “a dummy component made entirely of plastic.” It “contained no copper, and therefore had no conductivity.” It was marketed as a receiver, but “it was made of plastic and could not function as a receiver.” And it solved the fit problem perfectly, because it “could be cut to size by the doctor for use in smaller anatomical spaces.” A thing that does nothing can be trimmed to any length without loss of function.</p><p>The release is blunt about why it existed. The White Stylet was “created solely so that doctors and medical providers would continue to purchase the Device.”</p><p>The economics are in the same document. Stimwave sold the device to doctors for approximately $16,000. Implanting the Lead was reimbursable at roughly $4,000 to $6,000. Implanting a receiver was reimbursable at roughly $16,000 to $18,000. Then U.S. Attorney Damian Williams said doctors could bill Medicare and private insurers “approximately $18,000 for each implantation of that dummy component.”</p><p>Read that sequence again, because it is the mechanism. The fake was not built to fool a buyer into thinking they had bought something better. It was built so that a billing code would fire. The target was not the supply chain. It was the reimbursement schedule.</p><p>The scheme ran, per the release, “From at least in or about 2017 up to and including 2020.” Laura Perryman, Stimwave’s founder and then chief executive, was convicted after a two-week trial before U.S. District Judge Denise L. Cote of health care fraud and conspiracy to commit health care fraud and wire fraud, and Judge Cote sentenced her to six years in prison and three years of supervised release.</p><h3>Genuine by every test this series applies</h3><p>Now put the White Stylet through the checks.</p><p>Was it made by the company named on it? Yes. Stimwave made it. Did it ship through the authorised channel? Yes. Was it invoiced by the manufacturer of record, in the manufacturer’s packaging, with the manufacturer’s documentation? Yes. Would a serial number resolve to a real record in a real manufacturer’s database? Yes, because the manufacturer issued it. Would a cryptographic tag bound to the unit at the point of manufacture return a valid response? Yes, because the party doing the binding was the party doing the fraud.</p><p>Every identity mechanism this publication has described over thirty-nine articles returns the same answer on this part, and the answer is authentic. The answer is also correct. The part really was a genuine Stimwave White Stylet.</p><p>It was also a piece of plastic implanted in a human being under the representation that it was a receiver.</p><p>Here is the structural reason, and it is not specific to Stimwave. Every verification system in this field is a trust propagation system: it takes an assertion made at the point of manufacture and carries it, intact and tamper evident, to the point of use. The cryptography is excellent. What it protects is an assertion, and the party making the assertion is the manufacturer. The root of trust in every one of these architectures is the company that made the thing. When that company is the adversary, nothing downstream is compromised, because nothing downstream needs to be. The system works perfectly and propagates a lie with full integrity.</p><h3>This is not a rare failure</h3><p>It would be comfortable to file Stimwave as an oddity, one bad executive in one small company. The record does not allow that.</p><p><strong>Poly Implant Prothèse.</strong> The French manufacturer filled silicone breast implants with a gel other than the one it had declared. The EU’s own scientific committee found the company “fraudulently made use of low-quality material (industrial silicone)” and that the silicone differed from “the one it had declared in the documents submitted for conformity assessment.” The implants had been CE marked in October 1997 under the supervision of the notified body TÜV Rheinland, as the UK Department of Health’s 2012 report on the affair records, and the same report concludes that PIP “deliberately concealed their use of a non-approved filler material.” The European Parliament’s resolution of 14 June 2012 put around 400,000 PIP silicone implants sold worldwide, and said plainly that “the overall number of women who have received implants is unknown.” Jean-Claude Mas, the founder, was convicted of aggravated fraud in Marseille on 10 December 2013 and sentenced to four years, upheld on appeal in 2016.</p><p><strong>Takata.</strong> In the statement of facts attached to its 2017 guilty plea in the Eastern District of Michigan, the company admitted that from about 2000 to about 2015 it fabricated, removed and altered test data in validation and lot acceptance reports, concentrating on ballistic output and effluent gas, the two measures its ammonium nitrate inflators could not meet. Employees had a word for stripping out unfavourable results. They called it “XX-ing” the data. The DOJ’s fraud section chief said the company “repeatedly and systematically falsified critical test data related to the safety of its products.” Takata pleaded guilty to wire fraud and paid $1 billion, comprising $975 million in restitution and a $25 million fine. NHTSA’s recall page, as published and read on 9 November 2026, states that approximately 67 million Takata airbags have been recalled in the United States and that it has confirmed 28 deaths in this country.</p><p><strong>Ranbaxy.</strong> On 13 May 2013 the generic manufacturer’s US arm pleaded guilty to seven federal felonies, three under the Food, Drug and Cosmetic Act relating to adulterated drugs and four for knowingly making material false statements to the FDA, and the group paid $500 million. The false statements concerned stability testing for named antibiotics: tests reported to the agency as months apart had in some cases been run on the same day or within days of each other, and samples awaiting testing sat in a refrigerator that was not an approved stability chamber, which the FDA was not told.</p><p>Three different industries, three different decades, three different fraud mechanisms. One structural feature in common. In every case the product was made by the company whose name was on it, shipped through the authorised channel, and accompanied by correct and genuine documentation. Every provenance check in existence would have passed every one of them, and in PIP’s case a notified body audit actually did.</p><h3>Not one of them was caught by a check on the product</h3><p>Follow how each came to light, because the pattern is the point.</p><p>PIP was found by an unannounced regulatory inspection of the company’s premises on 16 to 18 March 2010, following a surgeon’s alert and a rise in device incident reports. Ranbaxy was brought down by a whistleblower, Dinesh Thakur, who received about $48.6 million of the federal recovery. Takata’s falsification surfaced through field ruptures, congressional scrutiny and a criminal investigation. And for Stimwave, as far as the public record goes, nobody has said at all: neither DOJ release names a whistleblower, a qui tam action, an FDA inspection or an internal audit, and the only physician reporting sentence in either document concerns the fit of the genuine part rather than the existence of the fake one.</p><p>Inspection, whistleblower, investigation, unknown. Not one of these four was caught by examining the product, scanning a code, checking a serial, or auditing a chain of custody. Every one was caught by a person with access, or by a body with subpoena power, or by bodies failing in the field. This is the single most uncomfortable fact in this article, and it is also the most useful one: the detection mechanisms that actually work against insider fraud are the ones nobody is selling, and the ones being sold would not have helped.</p><h3>And none of it counts as counterfeiting</h3><p>Now the part that makes this a measurement problem rather than an anecdote.</p><p>The headline figures everybody quotes for product fraud are trade based, built from cross border flows of infringing goods: the EUIPO and Commission detention data this publication reported on in October, the OECD’s estimates, customs seizure totals. We have used those numbers ourselves and they are the best available for what they measure, which is an outside party moving goods that infringe somebody else’s rights. A licensed manufacturer shipping its own out of specification product under its own name, through its own authorised distributors, infringes nobody’s trademark. It is not counterfeit in the legal sense that the measurement apparatus uses. So it is not merely under counted. It is outside the count by construction.</p><p>PIP appears in no counterfeiting statistic. Neither does Takata. Neither does Ranbaxy. Neither does the White Stylet. Four cases, at least 67 million recalled airbags, around 400,000 implants sold, $1.5 billion in criminal and civil resolutions between two of them, and a figure of zero in every chart the industry puts on a slide about the scale of the problem.</p><p>We went looking for any published estimate of how often product fraud originates with the authorised manufacturer rather than an outsider, across the OECD and EUIPO illicit trade work, the food fraud literature and the standard typologies. The typologies name insider origin fraud as a distinct category. None of them measures its share. As far as we can establish, no published estimate exists.</p><p>That absence is the reason this is not a niche concern. An entire regulatory and commercial apparatus, DSCSA, the EU Falsified Medicines Directive, digital product passports, the GS1 2D migration, national verification marks in Kenya and Pakistan and India, is being built to address a threat model that is measured, funded and understood. Running alongside it is a second threat model that produces some of the largest product safety events on record, that no instrument in the first apparatus can detect, and for which nobody has a number.</p><h3>Why the insider version is the cheaper attack</h3><p>Set the two side by side as an economic problem, because that is how the person doing it sees it.</p><p>An outside counterfeiter has to acquire the product, reproduce the packaging, forge the identifiers, build a distribution route, and keep all of it away from the brand, customs and the platforms. That is real capital and real operational risk, and it is why enforcement, for all its limits, has some purchase on it.</p><p>An insider changes a specification. The tooling exists, the staff are employed, the channel is already open, the customers already have purchase orders, the documentation generates itself, and every control in the chain is designed on the assumption that this party is telling the truth. There is no smuggling, no forgery, no parallel distribution, no border.</p><p>And the dwell times show it. Takata admitted roughly fifteen years of falsified data. PIP’s implants were on the market from 2001 and the substitution was not established until an inspection in 2010. Stimwave’s White Stylet ran from about 2017 through 2020. Outside counterfeiting rings get rolled up; insider fraud tends to run until somebody inside talks or something starts failing in the field.</p><h3>Nobody has published how many people have one</h3><p>Back to the specific case, and to an absence that should bother everybody who works on this.</p><p>Neither the indictment release of March 9, 2023 nor the sentencing release of June 17, 2024 states how many patients received a White Stylet. Not an estimate, not a range. Both say only that Perryman caused doctors and medical providers to implant it into patients, and Williams referred to “the patients who were implanted with that piece of plastic.” Neither release describes any patient’s outcome, and neither mentions explant or revision surgery. We could not find a published number anywhere.</p><p>Three years of implants at up to $18,000 a procedure, and the public record does not say how many people are walking around with a non-conductive plastic rod where a receiver was supposed to be. The European Parliament said the same thing about PIP in 2012: the number of women is unknown. Fourteen years apart, two insider frauds, the same missing figure, and in both cases it is the only figure a patient actually wants.</p><p>The FDA is almost absent from both Stimwave releases. Neither mentions the device’s clearance status, nor whether the White Stylet fell within whatever clearance the StimQ held. Trade reporting disagrees even on the clearance date, August 2017 against 2016, and we assert neither.</p><h3>What we are not saying</h3><p>We are not saying most product fraud is committed by the manufacturer of record. We have no basis for that claim and neither does anybody else, which is the point of the measurement section above. Outsiders copying somebody else’s product remain the most common case in this catalogue and in the enforcement record.</p><p>We are not saying identity verification is pointless. A mechanism that defeats outsiders is worth having against outsiders, which is most of the cases. The claim we are retiring is narrower and more important: that a verified unit is a safe unit. It is not. A verified unit is a unit from the party it claims to come from, and that is all it is.</p><p>We are not saying the doctors were at fault. The release says they were caused to implant it unwittingly, and the government’s case was that they were deceived about what the component did.</p><p>And we are not saying our own field would have caught this. It would not have. Not partially, not with better cryptography, not with a richer data carrier. Everything we build attests to origin, and the origin was real. This is the clearest case in thirty-nine articles of a failure mode our category cannot address, and a vendor who tells you otherwise is selling you something that does not exist.</p><h3>Who meets this problem</h3><p>The surgeon who asked a manufacturer whether a component could be cut to size, got a yes, and was telling the truth to the patient as far as anything available to them could establish.</p><p>The hospital value analysis committee whose supplier qualification file was complete, current and accurate the entire time.</p><p>The payer processing an $18,000 claim under a code for a receiver, with every upstream document in order.</p><p>The notified body or auditor who signed a conformity assessment against a technical file that described a product the company had stopped making.</p><p>The patient who consented to an implant and has no way to find out, today, which stylet they received.</p><p>And every quality lead who has been told, by somebody in our industry, that unit-level verification closes the authenticity question.</p><h3>If you are one of these people</h3><p><strong>If you buy implantable devices.</strong> Separate the two questions in your own paperwork. Provenance is answerable at receiving. Conformity is answerable only by specification, testing, or an independent party, and no amount of serialization substitutes for it. If a supplier’s answer to “does this part work” is “it is genuine,” that is a non sequitur, and it was the non sequitur at the centre of this case.</p><p><strong>If you are a payer.</strong> A reimbursement code is a target. The White Stylet existed because a code paid $16,000 to $18,000 for implanting a receiver and nothing downstream tested whether a receiver had been implanted. Codes that pay for a function rather than a procedure need a way to establish the function.</p><p><strong>If you run an audit or conformity assessment programme.</strong> PIP passed CE marking under a notified body’s supervision and concealed the filler from it. A document based audit against a technical file tests whether the file is consistent, not whether the factory is following it. Unannounced inspection and independent sampling are the controls that found PIP, and they are the expensive ones.</p><p><strong>If you are a regulator.</strong> The number of patients who received this component is not in the public record of a case that ended more than two years ago. That is a publishable number and it would be worth publishing. The same is true of the insider share of product fraud generally, which nobody appears to have measured.</p><p><strong>If you sell verification, including us.</strong> Say out loud where the product stops. Ours stops at the manufacturer’s own intent. A tag issued by the party committing the fraud verifies the fraud, and any claim that an authenticity layer makes a device safe is false. Sell it as what it is, a provenance control, and let people buy conformity controls separately, because they are a different product.</p><h3>What would have had to be provable</h3><p>For anybody to have caught the White Stylet at the point of care, the provable thing was never going to be where it came from. It came from Stimwave, and that was true all the way down.</p><p>What would have had to be provable, at the item, was what it contained and what it did. Whether a given rod carried a conductor. Whether the thing being billed as a receiver could receive. That is a conformity question, answerable by specification and by testing, and it belongs to a different discipline from the one we work in.</p><p>Which leaves the industry this publication belongs to with an uncomfortable summary, and it is worth stating without softening. We have built a decade of mandates, standards and products to answer whether a thing came from who it says. Four of the largest product safety frauds on record would have passed all of them, none was detected by any of them, none appears in any statistic measuring the problem, and nobody has counted how often it happens. The counterfeiter is usually outside. Occasionally the counterfeiter signs the certificate, and when that happens, every instrument we have built points the wrong way.</p><p><a href="http://aeroz.io/">Aeroz</a>, Making Authenticity Undeniable. Visit <a href="https://aeroz.io/technology">aeroz.io</a> to learn more &amp; get in contact with our team via info@aeroz.io.</p><p>2026 AEROZ all rights reserved.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a1fae11125b6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A Recall Cost Sage Tens of Millions. The Insurer Owed Nothing.]]></title>
            <link>https://medium.com/@aeroz.io/a-recall-cost-sage-tens-of-millions-the-insurer-owed-nothing-871e1c52202a?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/871e1c52202a</guid>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Thu, 05 Nov 2026 22:02:10 GMT</pubDate>
            <atom:updated>2026-11-05T22:02:10.905Z</atom:updated>
            <content:encoded><![CDATA[<p><em>A $13 million judgment, a $6 million claim, and an Illinois court holding in 2023 that none of it was an accident.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZH5W9wb3QpzswbKdpK4V0w.jpeg" /></figure><p><a href="http://aeroz.io/">Aeroz</a> <strong>Editorial</strong> November 2026</p><p>On June 29, 2017, the FDA told ChemRite CoPac that it had been making oral rinse for hospital patients on the same equipment it used to make toxic car wash. Sage Products recalled the kits and lost tens of millions. It thenspent six years finding out who pays for that, and guess what…the answer was nobody.</p><h3>The product and the warning letter</h3><p>Sage Products makes oral hygiene kits for hospital patients, sold as the Q-Care Oral Cleansing and Suctioning Systems. Each sleeve holds a toothbrush, a swab and one of several oral rinse solutions. Sage supplies most of the components itself and contracted ChemRite CoPac to manufacture and supply the rinse.</p><p>In the court’s recitation of the undisputed facts, the FDA’s warning letter of June 29, 2017 found that ChemRite’s manufacturing process “did not conform to relevant standards, including because ChemRite manufactured the solutions using the same equipment it used to make toxic car wash.” ChemRite told Sage the next day. On July 17, 2017, the FDA sent a warning letter to Sage.</p><p>Sage recalled and stopped distribution of everything containing ChemRite’s rinse. Federal guidelines required the rinse to be removed before disposal, which meant, in the opinion’s words, tearing open “each sleeve of each Q-Care Kit.”</p><p>The court records one figure and it is the only one it needed: “it is undisputed that Sage lost tens of millions of dollars in profits from the recall of its Q-Care Kits.”</p><h3>How Sage ended up suing an insurance company</h3><p>Sage sued ChemRite for breach of contract in August 2019. ChemRite asked its insurer, Federal Insurance Company, to defend it; Federal did not defend but paid for the defence. The parties mediated in late 2021, and Federal intervened in the case.</p><p>Then came the settlement that matters, and anybody who has watched a supplier dispute end will recognise its shape. Per the opinion, it “provided for a consent judgment of $13 million against ChemRite. ChemRite would pay $4 million, plus a contingent payment of $500,000 depending on Sage’s recovery from Federal, and ChemRite would assign its rights to insurance payments from Federal to Sage.”</p><p>So Sage, as the opinion puts it, “stepped into ChemRite’s shoes and sued Federal for $6 million, the maximum under ChemRite’s policy.”</p><p>A $13 million judgment on paper. $4 million actually paid. And the rest riding on one question about one word in a commercial general liability policy.</p><h3>The word was “occurrence”</h3><p>Federal’s policy obliged it to pay damages ChemRite became legally obligated to pay “by reason of liability” for “property damage caused by an occurrence to which this coverage applies.” The policy defines an occurrence as “an accident, including continuous or repeated exposure to substantially the same general harmful conditions.”</p><p>And then the gap that decides the case: as Judge Lindsay C. Jenkins noted in the Northern District of Illinois on October 25, 2023, applying Wisconsin law, “‘Accident’ is not defined in the policy.”</p><p>Wisconsin fills it from the case law. The means or cause must be accidental. The event must occur by chance, and be unforeseen and unintended. Faulty workmanship is not itself an occurrence, although faulty workmanship may cause one, a distinction the Wisconsin Supreme Court set out in <em>Walworth, LLC v. Engerman Contracting, Inc.</em> in 2023.</p><p>The court then asked the only two questions that could have produced coverage. Was putting the rinse into the kits an accident? Was taking it out one?</p><h3>Neither putting it in nor taking it out</h3><p>On incorporation, the opinion is blunt. Sage and ChemRite intended the rinse packets to go into the kits and be sealed, because “that was the purpose of their contract.” The court continues: “No one could say that the incorporation of the rinse solutions ‘occurr[ed] by chance’ or was ‘unforeseen and unintended’.”</p><p>On the recall, the same. It is undisputed, the court writes, that Sage recalled the kits, opened them and disposed of the rinse because government guidelines required it. “These were intentional and volitional acts, not accidents.”</p><p>The opinion then anticipates the obvious objection, that nobody chose any of this, that a federal agency forced the whole sequence. “The only difference with the situation here is that an outside party, the FDA, required the recall, but because those actions were intentional, not accidental, they do not amount to an occurrence.”</p><p>And it adds the sentence that should be printed and pinned above every supplier qualification file in the country:</p><p>“Nor can the fact that no one anticipated that ChemRite would provide a tainted product save Sage.”</p><p>Read that slowly, because it inverts what most people assume insurance is for. Your surprise is not the test. The question is not whether you knew the component was bad. It is whether what you did with it, putting it in and later pulling it out, was something you did on purpose. It was. Judgment for the insurer, case terminated.</p><p>The court never reached the exclusions. It did not need to. There was no grant of coverage to exclude.</p><h3>A second legal system reaches the same place</h3><p>Nine years on and eleven thousand kilometres away, the same reasoning turns up in a Chinese courtroom, and we flag immediately that our account of it is weaker than our account of Sage.</p><p>On July 9, 2026, lawyers at AnJie Broad published a note on what they describe as the first substantive Chinese ruling on the scope of coverage under a CGL policy. A manufacturer produced motor-shaft oil seals in 2020 for automotive electronic control drive units. An engineer changed process parameters without updating the process documentation, and a batch came out with what the note calls a serious durability defect. The manufacturer recalled. The customer sought roughly $4 million. The manufacturer claimed $2 million from its insurer, the single-occurrence property damage limit.</p><p>The insurer had already paid more than $1.5 million under a product recall extension clause, for dismantling and reinstallation labour and return shipping. What it refused was the value of the components scrapped in the process. Both the trial and appellate courts found there had been no occurrence, that the primary CGL coverage was never triggered, and that the manufacturer could recover only the agreed expenses under the recall extension.</p><p>Two caveats, and they are not small. The note was written by the insurer’s own counsel, who won the case. And it names no court, no case number and no decision date, so we cannot read the judgment or confirm it independently.</p><p>What it does show, if accurate, is that the structural answer is not an artefact of Wisconsin. The liability grant is built around accidents. A recall is a decision.</p><h3>What changes if the part was counterfeit</h3><p>Here is where we have to be careful, because it would be easy and wrong to say these are counterfeit cases. They are not. ChemRite was a contracted manufacturer with an FDA-regulated plant and a shared production line. The Chinese case was a process-control failure by the insured itself. Neither opinion mentions counterfeiting.</p><p>Our reading, offered as a reading and not as a holding, is that the reasoning is indifferent to how the bad part got in. The test the court applied was whether the acts were intentional, not whether the defect was known. Incorporating a counterfeit capacitor into a display assembly is as intentional as incorporating a tainted rinse into a kit, and more so, because you believed it was genuine and meant to use it. Pulling it back out under regulatory pressure is as volitional in Michigan as it was in Illinois. Nothing in that chain looks more like an accident because the supplier was a forger rather than a sloppy contract manufacturer.</p><p>And then the part that is worse, which took us by surprise when we worked through it.</p><p>Sage lost, but look at what Sage had. A named, solvent, identifiable supplier. A contract. A breach claim worth a $13 million consent judgment. $4 million actually collected. An insurance policy capable of being assigned, even if it paid nothing in the end. Every one of those is recourse, and every one of them exists because Sage knew exactly who made its rinse.</p><p>A counterfeit component offers none of it. There is no contract with the party that actually made the part. There is frequently no identifiable party at all, which is the finding this series has recorded again and again: NHTSA could not establish how many DTN60DB inflators are in the United States, and the falsified packs in Nigeria carried a manufacturer’s name whose connection to them nobody has established. You cannot sue a carton. You cannot take an assignment of a policy from a company that does not exist.</p><p>So the honest version of the lesson is sharper than the cases themselves. If the component is merely defective, you may lose the insurance argument and still have a supplier to pursue. If it is counterfeit, you may lose the insurance argument and have nothing to pursue at all.</p><h3>What we are not saying</h3><p>We are not saying that recalls are uninsurable. They are not. Product recall insurance is a distinct product with its own grant, and the Chinese case shows a recall extension clause doing exactly what it was bought to do: more than $1.5 million paid for dismantling, reinstallation and freight. The finding in both cases is narrower, that the general liability grant did not reach the loss. If recall exposure is covered, it is covered because somebody bought that cover on purpose.</p><p>We are not saying Federal behaved badly. It paid for ChemRite’s defence, it litigated a coverage question that a court agreed with, and insurers are entitled to the words they wrote.</p><p>We are not saying this is settled law everywhere. Sage is one district court applying one state’s law, and we found no published appellate decision in it, although we did not check the docket. Jurisdictions differ on faulty workmanship, and Wisconsin’s own position moved as recently as 2023.</p><p>And we are not saying our own field would have prevented any of this. Nothing we build would have caught ChemRite. The rinse was genuine product from the correct supplier, made on the wrong line, and an identity check would have confirmed every true thing about it. That is the same concession this series made about the Kota oxytocin deaths, and it holds here.</p><h3>Who meets this problem</h3><p>The quality lead who signs off a supplier qualification and does not know that surprise is not a defence in the coverage argument that follows.</p><p>The risk manager who has a CGL tower and no recall endorsement, and has been assuming the first covers the second.</p><p>The CFO who learns from a court, six years after the event, that the $13 million judgment on paper was always going to be $4 million in cash.</p><p>The brand protection lead whose counterfeit exposure is usually described in units seized, when the number that will reach the board is the cost of the pull and who absorbs it.</p><p>And the contract manufacturer with an honest process failure, who finds its own insurer and its own customer on the same side of the table against it.</p><h3>If you are one of these people</h3><p>If you buy the insurance. Ask your broker, in writing, whether a recall triggered by a regulator counts as an occurrence under your general liability wording, and what the recall endorsement actually pays for. The Chinese case is a useful checklist of what got paid: dismantling, reinstallation, return shipping. And of what did not: the value of the components destroyed in the process.</p><p>If you qualify the suppliers. The coverage question does not care what you knew. Build the file on the assumption that the loss lands on you and that your recourse is only as good as your ability to name and reach the party that actually made the part.</p><p>If you are a brand with counterfeit exposure. Price the recall, not just the seizure. Then ask who you would sue, and whether that party has assets, a policy, or a address. For a genuine supplier the answer is usually yes. For a counterfeit one it is usually no.</p><p>If you sell verification, including us. This is not a case our product wins. The right claim is narrow: knowing which units are affected and who supplied them decides the size of the pull and whether there is anyone to recover from. That is a cost argument, not a safety argument, and it should be made to a CFO in those terms or not at all.</p><h3>The bill is the deadline</h3><p>For six years, the single most consequential fact about this recall was not what was in the rinse. It was that two companies, a federal agency and an insurer all behaved exactly as they were supposed to, and tens of millions of dollars of loss still had nowhere to go but onto the company that assembled the kits.</p><p>What would have had to be provable, at item level, is narrower than the whole problem and worth stating precisely: which units contain the affected component, and who actually made it. The first bounds the recall. The second decides whether there is anybody left to pay for it. Sage had the second and still lost the argument. With a counterfeit part you can lose both, and the bill arrives anyway.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=871e1c52202a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Pakistan Will Seal Plants Over a Barcode. Old Packs Stay Exempt.]]></title>
            <link>https://medium.com/@aeroz.io/pakistan-will-seal-plants-over-a-barcode-old-packs-stay-exempt-183da9359f5a?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/183da9359f5a</guid>
            <category><![CDATA[barcode]]></category>
            <category><![CDATA[supply-chain]]></category>
            <category><![CDATA[track-and-trace]]></category>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Wed, 04 Nov 2026 21:16:05 GMT</pubDate>
            <atom:updated>2026-11-04T21:19:45.645Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Eighty per cent of companies had started by November 3. Every batch made before October 9 may still be sold.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5GLLq4Fv6FyQjRyoG-lo7Q.png" /></figure><p><a href="http://aeroz.io/">Aeroz</a> <strong>Editorial</strong> November 2026</p><p>On October 9, Pakistan began requiring a 2D Data Matrix barcode and a serial number on every medicine made or imported from that date. On November 3, officials put the figure at 80 per cent and the health minister said non-compliant plants will be sealed.</p><p>That is the fastest national pharmaceutical serialization rollout in recent memory, and the threat behind it is not idle. It is also, read closely, a mandate with a hole in the middle of it that nobody in the coverage has mentioned.</p><h3>The sentence at the end of the notification</h3><p>The requirement comes from <a href="https://www.dra.gov.pk/about_us/legislation/sros/notification-of-amendments-in-the-drugs-labeling-and-packing-rules-1986-09-06-2026/">S.R.O. 963(I)/2026</a>. Business Recorder described it as mandatory for “all registered drug products manufactured locally or imported into Pakistan on or after 9th October 2026.”</p><p>Then it quoted the clarification that follows:</p><p>“Further it is clarified that batches manufactured locally or imported with manufacturing date prior to 9th October, 2026 are exempted from this requirement and shall remain permitted for import, distribution, and sale.”</p><p>Nukta reported the same thing on October 12, 2026, in its own words: batches made or imported before that date “are exempt, DRAP said. They may continue to be imported, distributed and sold.”</p><p>As of October 9, there are two legally compliant categories of medicine on Pakistani shelves: packs manufactured on or after this date, which must carry a serialized Data Matrix code, and packs manufactured prior to it, which are exempt from this requirement and remain legal to import, distribute, and sell.</p><p>Neither report we could read gives an end date for the exemption. It runs, on its face, for as long as pre-October stock remains within its expiry.</p><h3>What that does to an inspection</h3><p>Here is the practical consequence, and it is the reason this piece exists.</p><p>For the length of that window, an unmarked pack is not evidence of anything. A drug inspector in Lahore, a hospital pharmacist in Karachi, a distributor’s goods-in clerk: each of them, holding a carton with no Data Matrix on it, is looking at something that is either lawful pre-October stock or a counterfeit printed to look like lawful pre-October stock. The absence of the code is compatible with both.</p><p>That is not a flaw anyone introduced carelessly. The alternative was to render every pack already in the country unsaleable on a single Friday, which in a market with real affordability pressure means shortages and destroyed legal inventory. Almost every serialization regime in the world has granted the same relief, for the same reason. The point is narrower: during the transition, the control cannot be used as a screen. It can only be used on the packs that already have it.</p><p>And the transition is where counterfeiters work. The racket does not need to defeat a Data Matrix. It needs to look like the stock that is exempt from carrying one.</p><h3>Eighty per cent of what</h3><p>The number from the November 3 review meeting does not say the same thing in every account of it.</p><p>The Nation, on November 4, 2026, reported that the meeting “was informed that around 80% of pharmaceutical companies had started implementing the 2D barcode system.” ARY News, on November 3, 2026, reported that “80 percent of pharmaceutical companies have already initiated the 2D barcode enforcement process.” 24 News HD, the same day, reported that companies “have begun implementing 2D barcodes on approximately 80% of medicines.”</p><p>Two of the three say 80 per cent of companies. One says 80 per cent of medicines. And two of the three attach the verb “started” or “initiated” rather than “completed.”</p><p>Those are three different facts with one number. Eighty per cent of firms having begun a project is not eighty per cent of packs carrying a verified code, and for anyone planning a market-entry check or a procurement rule, the difference decides whether the control is usable.</p><p>None of the three reports we read gives a count of how many companies or products that percentage refers to.</p><h3>The dates are not settled either</h3><p>This is where it gets harder to plan against, and we are reporting it rather than resolving it.</p><p>ARY News on September 24, 2026, reported that “the new requirements will apply to medicines manufactured from October 19, 2026, according to the notification.” The Destination, the same day, printed the same date in its subheading. Business Recorder on October 13 and Nukta on October 12 both give October 9. The two later reports agree, and The Nation’s November 4 piece refers to “the October 9 deadline,” so October 9 is the better-supported date. But the public record contains both.</p><p>The S.R.O.’s own date fares worse. Business Recorder’s October 13 report cites it twice in the same article: once as “S.R.O.963(I)/2026 dated 9th June, 2026” and once as “S.R.O.963(I)/2026 dated 9th July, 2026.”</p><p>More consequentially, two accounts of the notification say opposite things about unmarked stock. ARY News on September 24 reported that “after the stipulated deadline, medicines without barcodes will not be allowed to be sold in the market.” The Destination, the same day, printed the same proposition. Three weeks later, Business Recorder quoted the clarification exempting pre-October batches and permitting their sale.</p><p>Both statements describe the same notification. A compliance team reading the Pakistani press in good faith could reach either conclusion about whether an unmarked carton is saleable.</p><h3>Printing is mandatory. Checking is not.</h3><p>Now the part that matters most to anybody who has watched a serialization programme land somewhere else.</p><p>In none of the seven reports we read is there a national verification portal or app named, nor any statement that scanning is required of a distributor, a pharmacist or an inspector. Business Recorder’s October 13 report mentions “e-Services portal integration” and “portal deployment” in the context of industry training and data submission. The obligations described everywhere are to print and to submit: print the code on the pack, submit the serialization data to DRAP.</p><p>The November 3 meeting is consistent with that. Across all three reports, the minister’s direction is that companies submit their serialization data, and the sanction for failing to is that production units are sealed. The enforcement instrument points at the manufacturer’s filing, not at the pack on the shelf.</p><p>That is the normal order of construction and not a scandal. You cannot check codes that do not exist yet. But it does mean that at this moment the system can tell DRAP which companies have complied, and cannot tell a pharmacist whether the carton in her hand is real.</p><p>A trade report from June 30, 2026, on sustainabilitymea.com, described the eventual consumer-facing intent: a smartphone scan returning expiry date, usage instructions and the official retail price. That is a vendor-adjacent trade account rather than a government document, and we flag it as such. It is also worth noting what that list does and does not include. Expiry, instructions and price are label data. None of them is an answer to whether the pack is the pack the manufacturer issued.</p><h3>What we are not saying</h3><p>We are not saying the mandate is weak. Going from a cabinet decision in mid-2026 to a live national requirement in October, with eighty per cent of firms engaged within a month and the sealing of production units on the table, is faster than the United States or the European Union has managed on any comparable measure. Several much richer regulators have been talking about 2D migration for years.</p><p>We are not saying the exemption was a mistake. It is the humane choice and the standard one, and a regime that destroyed every pack in the country on day one would have cost patients more than the counterfeiters.</p><p>We are not saying anybody has been caught out. We have no evidence that counterfeiters have exploited the exemption window, and we are not suggesting they have. We are describing a gap in what can currently be checked.</p><p>And we are not saying our own field closes it on its own. On October 21, 2026, we reported DRAP’s own alert on five counterfeit batches found in Karachi, detected by a drug testing laboratory rather than by anyone looking at a pack. We could not re-open that source today. A verification layer answers whether a unit is the unit the manufacturer issued. It does not answer what is inside it, and for that a laboratory is still the only instrument.</p><h3>Who is exposed to this</h3><p>The drug inspector working the transition window, for whom a missing code is not a finding.</p><p>The hospital pharmacist whose goods-in check cannot distinguish exempt stock from a forgery of exempt stock.</p><p>The compliance lead at a multinational selling into Pakistan, trying to pin a deadline that the public record states as both October 9 and October 19.</p><p>And the Pakistani manufacturer that has filed its serialization data and printed its codes, now competing on a shelf where the absence of a code carries no penalty for two populations of pack, only one of which is legitimate.</p><h3>What would have had to be provable</h3><p>For the exemption window to be a safe place rather than an open one, the question an inspector asks would have to be answerable at the item rather than at the carton. Not “does this pack carry a code,” which during the transition distinguishes nothing, but “is this unit the unit its manufacturer released,” answerable on the pack itself, by someone standing in front of it, without a reference database that is still being deployed and without the cooperation of whoever printed the label. That is an item-level identity bound at the point of manufacture, checkable in a second by the person holding the box. Pakistan has mandated the carrier for it in under five months, which is the hard part and genuinely impressive. What is not yet built is the check.</p><p><a href="http://aeroz.io/">Aeroz</a>, Making Authenticity Undeniable. Visit <a href="https://aeroz.io/technology">aeroz.io</a> to learn more &amp; get in contact with our team via info@aeroz.io.</p><p>2026 AEROZ all rights reserved.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=183da9359f5a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[100,000 People Came Looking for Counterfeit & Supplychain Optimization This Last Month]]></title>
            <link>https://medium.com/@aeroz.io/100-000-people-came-looking-for-counterfeit-supplychain-optimization-this-last-month-bf1505d5beeb?source=rss-e00a2bca2ed7------2</link>
            <guid isPermaLink="false">https://medium.com/p/bf1505d5beeb</guid>
            <category><![CDATA[supply-chain]]></category>
            <dc:creator><![CDATA[Aeroz]]></dc:creator>
            <pubDate>Tue, 03 Nov 2026 18:57:57 GMT</pubDate>
            <atom:updated>2026-11-03T18:57:57.954Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Nine countries. A grey area in the supply chain is being named, and the regulation is already on the calendar.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dIqWRIKGgiL_gmb2Ix6iqw.jpeg" /></figure><p><a href="http://aeroz.io/">Aeroz</a> <strong>Editorial</strong> November 3rd</p><p><strong>In the last 30 days</strong>, aeroz.io had 100,000 unique visitors across nine countries. Unique visitors, measured with bot filtering on, because a number worth reporting is worth stating precisely.</p><p>One hundred thousand people went looking for an answer to a question that has started arriving in a lot of jobs at once: <strong>can I prove what this product is?</strong> They found us working on it.</p><p>That is the signal. A gap that has sat unnamed in the supply chain for decades is being recognised as a distinct problem with a distinct answer, and the people whose job it will be to solve it are reading up before they have to.</p><h3>The grey area now has a name</h3><p>Between what the record says about a product and what the object actually is, there has always been a space nobody owned.</p><p>Procurement owns the contract. Quality owns the specification. Logistics owns the movement. Compliance owns the paperwork. Brand protection owns the trademark. And the question of whether the unit in somebody’s hand is the unit the paperwork describes has sat between all five of them, belonging to none.</p><p>That is the grey area. It is why a serial number can be genuine and the pack still falsified, why a carton can pass every data check and fail an artwork comparison, why a licensed factory with correct records can ship a vial with nothing in it. Every one of those is a documented 2026 case, and every one of them happened in the space between the record and the object.</p><p>A hundred thousand people arriving in a month is what it looks like when a space like that stops being invisible.</p><h3>The regulation is already dated</h3><p>This is the part that makes the interest a leading indicator rather than a mood. The rules are not speculative. They are written, numbered and scheduled.</p><p><strong>The European Union</strong> cited six digital product passport standards as harmonised standards in the Official Journal on 15 July 2026, under Implementing Decision (EU) 2026/1736, with two more on data security and authentication published in September. The battery passport becomes mandatory on 18 February 2027, with an item-level unique identifier and the obligation falling on whoever places the battery on the EU market.</p><p><strong>EU customs</strong> completed its largest overhaul since 1968. Regulation (EU) 2026/2108 entered into force on 21 September 2026. The €150 duty exemption is gone, a €2 handling fee started on 1 November, non-EU platforms are now importers for distance sales and liable for product compliance, and the Customs Data Hub becomes mandatory for e-commerce on 1 July 2028 and for all importers on 1 March 2034.</p><p><strong>The United States</strong> measures small-dispenser status under the drug supply chain rules on 27 November 2026, with the exemption running to 27 November 2027, and food traceability under FSMA 204 lands in July 2028.</p><p><strong>GS1</strong> migration to 2D carriers begins in Europe in early 2027 and the United States later that year.</p><p><strong>And it is not only the big blocs.</strong> Kenya has announced a national anti-counterfeit security device under section 34B of its own Act. Korea’s national mint is printing a verification mark for export goods. Thailand is building track and trace into its pharmaceutical rules. Maharashtra has put laser spectrometers into flying squads to test medicines in the field.</p><p>Four continents, inside eighteen months, all arriving at the same question from different directions.</p><h3>Who is arriving early</h3><p>Four kinds of reader, as far as the traffic shows.</p><p><strong>People with a date.</strong> The 2027 and 2028 deadlines above are now close enough to budget against, and they arrive wanting to know precisely what a rule requires.</p><p><strong>People who have had an incident.</strong> Something turned up that was not what it claimed to be, and they are working out what can be checked next time.</p><p><strong>People doing diligence.</strong> They want to understand what a verification mechanism establishes before they buy one, which is exactly the right order to do it in.</p><p><strong>People paying attention early.</strong> The ones who can see a compliance obligation forming and would rather meet it on their own schedule than someone else’s.</p><p>Healthcare and pharmaceuticals lead by a distance. Automotive is second. That is also the order in which the regulation is heaviest and the harm is best documented.</p><h3>What we take from it</h3><p>Interest at this scale, across nine countries, pointed at a problem that now has dated regulation behind it in four continents, is market formation. It is the early part of it, and it is unmistakably that.</p><p>What it tells us is that the argument no longer has to be made from first principles. We are not explaining why product identity matters to people who have never considered it. We are answering a question the reader brought with them, usually with a deadline attached.</p><p>So the work continues exactly as it is. One named case at a time, read from primary sources, with every figure attributed and every date verified. That is what the hundred thousand came for, and it is what is worth giving them.</p><p><a href="http://aeroz.io/">Aeroz</a>, Making Authenticity Undeniable. Visit <a href="https://aeroz.io/technology">aeroz.io</a> to learn more &amp; get in contact with our team via info@aeroz.io.</p><p>2026 AEROZ all rights reserved.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bf1505d5beeb" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>