<?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 Paul Chaumont on Medium]]></title>
        <description><![CDATA[Stories by Paul Chaumont on Medium]]></description>
        <link>https://medium.com/@paul.chaumont?source=rss-ec3e3a9f2193------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*CwwWn5WTwJFsdYgHnVh_vA.jpeg</url>
            <title>Stories by Paul Chaumont on Medium</title>
            <link>https://medium.com/@paul.chaumont?source=rss-ec3e3a9f2193------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 04:58:40 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@paul.chaumont/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[Contextual Stories — Chapter 3]]></title>
            <link>https://medium.com/context-insights/contextual-stories-chapter-3-98e4c260f4de?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/98e4c260f4de</guid>
            <category><![CDATA[medium]]></category>
            <category><![CDATA[computer-vision]]></category>
            <category><![CDATA[online-marketing]]></category>
            <category><![CDATA[contextual-marketing]]></category>
            <category><![CDATA[video-marketing]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Tue, 10 Aug 2021 12:49:06 GMT</pubDate>
            <atom:updated>2021-08-10T12:49:06.209Z</atom:updated>
            <content:encoded><![CDATA[<h3>Contextual Stories — Chapter 3</h3><p><em>Discover Contextual Stories, a tale of contextual advertising in 2021, a critical year for the media industry !</em></p><p><em>This third chapter focuses on the online video industry and how it has drastically changed over the last decade, opening new territories for all stakeholders</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bLNbklUIaN_wWu8sWUVVFw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PLa6hiVYEjZn7PKantWrbg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1heZotSDQpYJCJkSeZe43A.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6FrYpKBn-2dlf2aDq5pPRg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2MC6LfBmUEOwo1hwm_pweg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CygaKxH1BNFqBf5ak-YZSA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JSzV1EkD3PKvECoBXCMpSA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bnVZDpgz3H5uHC5yxRQkSA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4mreWjOUvt8s4G3-stkwDw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BdcWaPGXXjxHV9BT6EJKDw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D_sM-PG9ONTTl2ZrQ2b4og.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*l464ED01QXhDLWe8Ct7cjQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LNn0i6IG0RZ-5Psv-kNv5Q.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BUBgDrLiD6knzZDEn9sYGw.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=98e4c260f4de" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/contextual-stories-chapter-3-98e4c260f4de">Contextual Stories — Chapter 3</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Context presents Reach, 
the new standard for Contextual Advertising and Brand Safety]]></title>
            <link>https://medium.com/context-insights/context-presents-reach-the-new-standard-for-contextual-advertising-and-brand-safety-c78b93f728d8?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/c78b93f728d8</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[brand-safety]]></category>
            <category><![CDATA[contextual-marketing]]></category>
            <category><![CDATA[publishers]]></category>
            <category><![CDATA[advertising]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Tue, 25 May 2021 11:47:38 GMT</pubDate>
            <atom:updated>2021-05-25T11:51:42.720Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>Context presents Reach, <br>the new standard for Contextual Advertising and Brand Safety</strong></h3><p>Context is thrilled to present Reach, its brand new Contextual Advertising solution. Reach is a platform that enables creating a contextual advertising campaign based on video inventories matching brands’ contextual environment, from broad level to surgical precision. Reach is entirely customizable and will always adapt to your needs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1014/1*HF9aD-LnpBn2vWR72XAZ4A.png" /></figure><p><strong>Context processes millions of video from top world publishers at the speed of light.</strong></p><p>Using Reach, brands now launch contextual campaigns in a snap benefiting from videos’ momentum to build engagement with strong conversion KPI’s as a result without cookies. All of it guaranteed in a brand suitable environment thanks to the best brand safety vision technology available on market.</p><p>Based on your selection of IAB classification standards and GARM Brand Safety Categories, it ensures both the best context of appearance and the safest places. Include topics of interests, exclude sensitive topics and select top publishers amongst thousands to give the ideal context to your campaign.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7htTQMg_2tIt9NV-9Ixsww.png" /></figure><p>Built to provide the smartest advertising tool on video content, Reach now ensures your ads run in the right place, at the right time, with the right mind.</p><p><a href="https://wearecontext.com/reach/">Discover Reach right now</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c78b93f728d8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/context-presents-reach-the-new-standard-for-contextual-advertising-and-brand-safety-c78b93f728d8">Context presents Reach, 
the new standard for Contextual Advertising and Brand Safety</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Can we finally trust Artificial Intelligence on Brand Safety matters?]]></title>
            <link>https://medium.com/context-insights/can-we-finally-trust-artificial-intelligence-on-brand-safety-matters-7920ecc580b6?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/7920ecc580b6</guid>
            <category><![CDATA[social-media-marketing]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[digital-marketing]]></category>
            <category><![CDATA[brand-safety]]></category>
            <category><![CDATA[brand-strategy]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Fri, 22 Jan 2021 09:21:26 GMT</pubDate>
            <atom:updated>2021-01-22T09:21:26.150Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Brand Safety applied to video media is a topic that we have been discussing many times already on this blog and that has become key to the digital advertising conversation. In 2019, Alphabet was already spending hundreds of </em><a href="https://www.reuters.com/article/us-alphabet-google-youtube/google-spends-hundreds-of-millions-of-dollars-on-content-review-letter-idUSKCN1S81OK"><em>millions of dollars annually on content review</em></a><em> only to moderate about 300K terrorist videos monthly on YouTube. A very high cost for small results.</em></p><p><em>As time goes by and human moderation seems (only seems) to slowly fade away, one could think that Artificial Intelligence is now ready for a complete takeover. But is it truly? And most importantly, what can we really put behind Brand Safety, a catch-all term that many companies thrive on without having the real technologies to provide?</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n2OoODUoCmdWUmuCcW7wEg.jpeg" /></figure><p><strong>Brand Safety, Brand Suitability or Brand Equity?</strong></p><p>When addressing the matter of Brand Safety, people commonly use different acceptances that refer to specific meanings, thus services. Three should be taken into consideration in the context of Brand Image and content moderation : Brand Safety, Brand Suitability and Brand Equity.</p><p>Brand Equity refers to the value of a brand, determined by the consumer’s perception. It can either be positive or negative and affect certain features of your brand such as quality, credibility, or consideration.</p><p>Brand Safety refers to all practices ensuring that any advertising campaign does not appear linked to any dangerous or illegal content. This mostly includes content featuring violence, hate speech, drugs, alcohol or any form of nudity and sex.</p><p>Brand Suitability refers to all practices ensuring that any advertising campaign does not appear linked to any content that would be judged irrelevant or inappropriate for the brand. This more broadly acceptance is often summed up by saying : if it is aligned with your brand image, guidelines and strategy, it is brand suitable.</p><p>Once that said, we could sum it up as follows : regarding digital content, Brand Suitability ensures a brand to appear in a relevant and appropriate content, including Brand Safety matters, to protect or improve its Brand Equity. As <a href="https://www.adcolony.com/blog/2019/04/12/2019-update-how-does-brand-safety-impact-consumer-perception/">Facebook (60%) and YouTube (31%) remain the top places where consumers face non-brandsafe content</a>, negative consumer perceptions are now a key topic for brands : 50% of social media users <a href="https://www.marketingdive.com/news/marketers-unify-to-stop-flow-of-ad-revenue-to-harmful-content/571003/">admit having changed their opinion of a brand whose ads were displayed next to inappropriate content</a>.</p><p><strong>A single classification to rule them all?</strong></p><p>As Brand Suitability can vary a lot from one brand to another, we will limit the following to Brand Safety matters. The main issue then, is to agree upon a common framework to determine what is safe and what is not and how to introduce degrees in terms of severity. Actors such the Global Alliance for Responsible Media or Integral Ads Science have issued specific classifications and taxonomy to try to set limits to what could be deemed as unsafe. Solution providers such as AWS also provide their own classification. Most of them undergo rigorous auditing and are accredited by the Media Rating Council, giving them credibility on the advertising market.</p><p>But all of them still face two major issues preventing them from going towards a universal Brand Safety taxonomy. First, most taxonomies slightly consider video format and mostly favor a broad and common acceptance of media format, mixing text, image, website origins and in some cases voice. This explains why categories such as “Adult &amp; Explicit Sexual Content” go along “Spam” or “Online Piracy”. In the case of video media, these classifications are most times irrelevant : they provide too many details on useless categories and too few on major ones.</p><p>This is the second issue regarding existing classification : they all define their own level of sensibility regarding major topics such as “terrorism” or “gore content” but these are fine lines that would truly benefit from a common acceptance by all parties. Regarding topics such as nudity, once again, what is deemed acceptable or not is up to the country, the culture of the brand and more than anything : the context. Many topics that, in the case of video, are still not really considered.</p><p><strong>Understand video like humans do</strong></p><p>Thus, understanding the context of appearance becomes key in Brand Safety and this explains why video understanding automation is the ultimate tool for media stakeholders. A fake terrorist in a comedy could easily dress exactly like a reel one but would this video content be considered as apology for terrorism? Whether nudity is displayed in a serious documentary or an erotic movie can completely change our perception of nudity and its severeness. As for weapons, footage of a school shooting could not be put at the same level as displaying military troops equipped with guns marching during a parade. It’s all about context of appearance.</p><p>This is where the game becomes tricky. As of today, some companies do a fairly good job using A.I to analyze semantic, image or sound and get a fair assessment of the context. Video is the next level as it requires combining many various kinds of learnings, from the easiest ones (a plastic gun used by a kid is not a real gun) to the toughest cases (one could as well drink whisky in a water glass and water in a wine glass, how to make the difference?).</p><p>Let us take a single example of textbook Brand Safety to understand issues that are raised with video and context : promotion and advocacy of alcohol use. Display of alcohol bottles and products are rather easy to identify. Most times, shapes, logos, and colors really help the algorithm detect what is alcohol and what is not. But when it comes to drinking in other containers, behavioral understanding is necessary : drinking shots at a party, playing a beer game on a ping-pong table, or simply pouring translucid alcohol in a plain glass are as many examples as possible that require understanding that context is most likely to be displaying alcohol. Identifying someone being drunk without even identifying alcohol creates new difficulties, based on body movement, facial expressions, voice, overall context.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PVqtiyfEdNG2bEcAwc-RQA.jpeg" /><figcaption>Alcohol identification is not just a matter of detecting beer glasses</figcaption></figure><p>In the end, it all comes down to levels of understanding, as we humans do. Eventually, algorithms can help brands make assumptions on the context of a video, thus helping them assess the true necessity to reject a video.</p><p><strong>Towards levels of safety and customized variations</strong></p><p>In theory, Brand Safety applied to video is a quite simple affair : reject weapons, reject nudity, reject drugs and alcohol. But practically, once you block the “simple” cases, you are left with a vast majority of “grey area” cases for which Brand Safety becomes relative or subjective.</p><p>Again, let us take a simple example to understand it: the events that occurred to the U.S capitol a few days ago. Most videos from this event display weapons, violence, street riots, hate flags and signs. Whether you judge them Brand Safe or not will depend on many factors : the level of violence displayed, the country it airs, the handling of information itself or even the context of publishing. All this will have an impact on how confident a brand is to be associated with such videos.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gjRcMOM-gBSY47MPamITMQ.jpeg" /></figure><p>Brand Safety applied to video cannot simply be limited to a narrow classification. Any reasonable solution and offering real A.I powered brand safety monitoring (And not pretending to do so by hiring external human moderation partners) should indeed include two major cornerstones in their approach :</p><ul><li>Always give probabilities of content safety and not strict answers. Sometimes it can be 100% safe or unsafe. But most times, it will be way more realistic to provide a tendency rather than an assertion.</li><li>These tendencies must rely upon the brand’s own guidelines and acceptance towards any topic related to Brand Safety. And even for a single brand, these should be adaptable to local constraints. Nudity appearances for instance should rely upon customizable levels depending on needs and constraints.</li></ul><p>In the end, Artificial Intelligence helps us provide a rational understanding for a subjective interpretation. This subjectivity is each one’s to set but our role as A.I solution providers is to give them the means to make their own choices.</p><p>There are still many complexities related to video understanding and we at Context work each day to increase our client’s capabilities to define their Brand Safety (and suitability) guidelines. The final argument for the advent of A.I applied to Brand Safety is that it will open a new era for programmatic advertising : being able to monitor pre-bid at scale and increase speed of response. That is what will eventually make the difference between manual moderators (and A.I wannabes), mostly offering post-bid reports, and real A.I-powered Brand Safety solutions, such as the one we offer at Context.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7920ecc580b6" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/can-we-finally-trust-artificial-intelligence-on-brand-safety-matters-7920ecc580b6">Can we finally trust Artificial Intelligence on Brand Safety matters?</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DO MASKS HAVE A REAL IMPACT ON FACIAL RECOGNITION ?]]></title>
            <link>https://medium.com/context-insights/do-masks-have-a-real-impact-on-facial-recognition-34c6cc52ae76?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/34c6cc52ae76</guid>
            <category><![CDATA[facial-recognition]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[covid19]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Tue, 24 Nov 2020 08:23:17 GMT</pubDate>
            <atom:updated>2020-11-24T08:23:31.041Z</atom:updated>
            <content:encoded><![CDATA[<h3>DO MASKS HAVE A REAL IMPACT ON FACIAL RECOGNITION ?</h3><p><em>Since Covid-19 has slowly become central in our daily lives, we all (or almost) got used to wearing a mask pretty much everywhere outside our houses. To convey this message, public figures have had to play their part as well and wear masks on media appearances. But what is the impact of wearing a mask on applied facial recognition ?</em></p><p><strong>BE MASKED OR NOT BE MASKED, THAT IS THE QUESTION</strong></p><p>November US presidential elections could not be a better example of two different approaches regarding masks on public appearances. It is no surprise that Donald Trump has appeared way less with a mask than Joe Biden. This led our team to wonder if such a difference would have an impact both on recall and precision of our results. Based on 450K minutes of videos processed by our algorithm during the US election, Context has analyzed variations on both candidates to compare results with and without a mask.</p><p>It appears that surprisingly enough, the presence of a mask is never a sufficient criteria to challenge our confidence threshold to identify both candidates (and we’ll explain why below) but the confidence score itself is more affected for Donald Trump than Joe Biden. In other words, when Donald Trump wears a mask, our algorithm slightly lowers its confidence score as it is almost unchanged for Joe Biden. A first explanation would be that Joe Biden’s upper facial part is more distinctive than Donald Trump’s. Another one could be that Trump’s mask being bigger than Biden’s and covering a wider fraction of his face, it becomes harder to recognize him.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IOHKMqSrnIKRSeV8LTxnBQ.jpeg" /><figcaption>Joe Biden’s mask uncovers more of his face than Donald Trump’s and could have an impact on recognition</figcaption></figure><p>In both cases, performances are slightly lowered but the overall system and recognition result is resilient. Why is that?</p><p><strong>A FACE IS SUM OF INFORMATION</strong></p><p>To assess this question, it is essential to understand how the algorithm learns : it does not learn to recognize Donald Trump or Joe Biden based on pictures as a whole but on infinite small criteria on each picture. Throughout this process, the network will make a complete use of any part of each picture from the database to refine its learning. To say it simply, it does not only learn based on the shape of the mouth or size of the nose but on many other criteria beyond basic human points of interest. The distribution of criteria improves the system’s resilience. A face becomes a sum of information, for which only a small fraction of it becomes necessary to ensure clear results. Whether the face is hidden by a mask, sunglasses, a cap or even from side view, the algorithm still has enough information to provide the right answer.</p><p>The second reason why masks did not affect that much our results is the amount of data our system is trained with. The bigger the dataset, the more comparison points you get, the more the algorithm gets to be trained on various cases and compensates when part of data is lacking.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pIZowLHX_2FPI0JEIZ2G3g.jpeg" /></figure><p>In cases when your own data is insufficient, one could think that adding pictures of the celebrity wearing a mask to the database could be a good idea to improve recognition in such a situation. But is it really?</p><p><strong>PROVIDING MASKED PICTURES TO IDENTIFY MASKED PEOPLE ?</strong></p><p>As it would seem the most logical solution, it is actually the riskiest. Adding pictures of specific people wearing masks to the database increases the risk of introducing what we call a wolf in the database. A wolf is an identity that manages to impersonate another identity in the database. Wolves are successfully imitating other identities and leading to increase false positives in results, one could call them impostors.</p><p>What is the impact on results then ? Having just a few masked pictures in your database will increase the chances that any presence of a mask in a video content you’ll process will always give the same celebrity result : the one that has pictures with masks in the database. It is a current problem one could face with celebrities wearing caps or sunglasses if these are not distinctive and always-worn items. A good example of this is the tricky case of guitarist Slash : does wearing sunglasses and a top-hat make you Slash ? And what about cases when / if Slash does not wear both items in public ? Will the AI be able to identify him ?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D2w5Sx8tr1DP2BvmEZWSwg.jpeg" /><figcaption>Sunglasses and Top-hats and clear markers to identify slash : even more than his face ?</figcaption></figure><p>On can then understand that the non-systematic presence of masks in datasets is an issue. As the algorithm has never been trained on masks, the mask is not an identity marker per say but the network never learnt to ignore it and to consider it as anecdotal.</p><p>To make this approach efficient enough, one would probably have to systematize adding masked pictures on all celebrities (or a fair amount) of the database. This would be the only way to “ignore” the masks.</p><p><strong>THE EFFORT IS NOT WORTH THE RISK</strong></p><p>Overall, taking the chances to enrich your database with masked pictures can only be considered if your original results are good enough. If for a masked Joe Biden, the algorithm does not even provide Joe Biden as the main answer, adding masks to the database will not change the game. In other words, such a solution can only improve results that are already good, not turn bad ones to excellent ones.</p><p>Another risk has to be highlighted regarding adding pictures of masked public figures: personalized masked. Let us take an example of a political party branded mask : adding such a mask to a celebrity’s database, Donald Trump’s for instance, creates a risk if any random people, let’s say his supporters, wear the same mask in a video. The singularity of the mask, what makes it different from others, becomes the reason why it is misleading. For this reason, any addition of masked pictures in your database would probably have to be neutral-enough so that it does not create misleading hints on who is behind this mask.</p><p>In any case, all the elements present above led us to think that the effort of adding masked pictures to the database was not worth the risk, especially when our algorithm was able to maintain enough confidence in the results at this stage.</p><p>Last month, Andrew Maximov, a Belarusian technologist living in Los Angeles posted a video on YouTube showing how facial recognition technologies could actually<a href="https://www.nytimes.com/2020/10/21/technology/facial-recognition-police.html"> remove masks from faces in images or video content</a> and retrieve the person behind, through social networks profiles. Although it is not proved at this stage that such matches are accurate, this case in point shows that use of masks did not put an end to facial recognition efficiency.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=34c6cc52ae76" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/do-masks-have-a-real-impact-on-facial-recognition-34c6cc52ae76">DO MASKS HAVE A REAL IMPACT ON FACIAL RECOGNITION ?</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Reminiz announces name change to Context]]></title>
            <link>https://medium.com/context-insights/reminiz-announces-name-change-to-context-f433c49088b8?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/f433c49088b8</guid>
            <category><![CDATA[branding]]></category>
            <category><![CDATA[advertising]]></category>
            <category><![CDATA[digital-marketing]]></category>
            <category><![CDATA[video-marketing]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Tue, 15 Sep 2020 07:57:21 GMT</pubDate>
            <atom:updated>2020-11-24T08:24:16.476Z</atom:updated>
            <content:encoded><![CDATA[<h3>REMINIZ ANNOUNCES NAME CHANGE</h3><p><em>Nearly 6 years after having started as a world pioneer in video understanding, Reminiz has announced today that the company will operate under a new name and be known as </em><strong><em>Context</em></strong><em>.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/528/1*uIQnzQ5LdjBd1Nhh5aH2MA.png" /></figure><p>This change is part of the startup’s global evolution, offering increased ai-powered technologies and a wider range of services. This is being followed by a new website and a brand-new identity giving form to <strong>Context</strong> new ecosystem.</p><p>The flow of daily uploaded videos has become humanly uncontrollable, leading to high levels of dubious videos visible by the masses. More than ever, <strong>Context</strong> serves the video industry as the ultimate video understanding A.I solution, solving two core problems regarding video content : content safety and content monetization. <strong>Our fully automated solution indexes and categorizes any video at scale to help uplifting video inventories value</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JeIYYERZClFpG3ePUltDZQ.png" /></figure><p>Based on proprietary and state-of-the-art algorithms, Context is the only video-ready solution with complete video-understanding capabilities, empowering the whole industry, from advertisers to publishers. It has developed fully automated processing pipe at scale on facial recognition, object detection, logo identification, sports and landscapes recognition, explicit content and violence detection, and is already well advanced on new and complex verticals.</p><p><strong>Context</strong> provides precise and complete metadata and ensures a granular analysis for three core services : contextual targeting, brand safety and content management.</p><p>The company is now excited to enter a new era of its development and is eager to discuss upcoming and innovative projects with existing and new partners.</p><p>To learn more : <a href="http://www.wearecontext.com/">www.wearecontext.com</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f433c49088b8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/reminiz-announces-name-change-to-context-f433c49088b8">Reminiz announces name change to Context</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TRUST, DATA SECURITY & INNOVATION — A BRIEF LOOK AT TOMORROW'S NEWS REPORT]]></title>
            <link>https://medium.com/context-insights/trust-data-security-innovation-a-brief-look-at-tomorrows-news-report-12ece2208b9e?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/12ece2208b9e</guid>
            <category><![CDATA[news]]></category>
            <category><![CDATA[consumer-behavior]]></category>
            <category><![CDATA[content-marketing]]></category>
            <category><![CDATA[advertising]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Tue, 11 Aug 2020 08:58:51 GMT</pubDate>
            <atom:updated>2020-08-11T09:11:36.007Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>TRUST, DATA SECURITY &amp; INNOVATION — A BRIEF LOOK AT TOMORROW&#39;S NEWS REPORT</strong></h3><p>For the fourth consecutive year, Reuters has published its <a href="https://www.reutersagency.com/en/reuters-community/tomorrows-news-2020-survey/?utm_source=Eloqua&amp;utm_medium=email&amp;utm_campaign=&amp;utm_content=Media%20-%20Triggered%20Email%20-%20Tomorrow%27s%20News%202020">Tomorrow’s News report</a>. The study was conducted over 2 months, December 2019 to January 2020, on over 1,700 Reuters professionals working in various fields such as IT, Finance, Media/Marketing, and Manufacturing amongst others. The insights that it brings on the evolution of online use and news consumption leads us to imagine what strategic levers will rule over the internet tomorrow.</p><p><strong>THE VALUE OF TRUST</strong></p><p>It is perhaps the most prominent topic that comes out of the survey and yet the most subject to debate. The value of trust, when it comes to news consumption is a key driver for users to select their sources. As a result, 72% of respondents “<em>selected ‘trusted and intelligent content’ as a factor that makes a news brand appealing</em>”. To the contrary, social medias rank last (32% of respondents) when it comes to be the top online source for news.</p><p>Traditional media, through their online platforms, still have a clear advantage over pure players : they remain symbols of accuracy, impartiality and trust compared to less organized or professionalized media. It is however important to put this in context of how media consumption shifted from a text-oriented approach to a video-oriented approach over the years.</p><p>As video is set to represent 90% of worldwide internet traffic in 2022, video content has become a strategic lever for media brands <a href="https://medium.com/reminiz-insights/content-monetization-at-the-lockdown-era-nows-the-time-to-play-it-smart-8f1267c223ce?source=collection_home---4------1-----------------------">to reinvent themselves </a>and preempt new territories such as social platforms. New short featured formats have created innovative ways to get information in a snap through trusted sources.</p><p>In the meantime, personal video content outbreak and editing technologies available for all have greatly diminished the value of User Generated Content overall : even though very popular for some, UGC related to news consumption is often associated with biased information, untrusted sources and lack of judgement. “64% of people have a more favorable opinion if a news appears on a trusted news site”, Reuters says.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/832/1*DuenNu1Y49wwPX-szCB9pw.png" /><figcaption>Credits — Tomorrows News Report’s from Reuters</figcaption></figure><p>With the amount of time people spend on social media every day, estimated to be around 3 hours in Western countries, it could be thought that news on social media would be taken more seriously. Yet interestingly enough, it is used as a complementary source to other news and not as the primary source of information. It is through social media however that the highest amount of video formatted content is shared, whether it be regular video or live streams.</p><p>This brings two important topics regarding providing valuable video content on media platforms : the ability for standard media to provide only and strictly only verified, valuable content and filter everything that would not follow strict brand safety guidelines. On the other hand, the necessity to clean social media from unverified, harmful or even worse dangerous video content, for them to become reliable sources for the audience. To this aim, Reminiz provides such capability and ensures full safety of all distributed content, no matter the medium.</p><p><strong>DON’T MESS WITH PERSONAL DATA</strong></p><p>The everlasting preoccupation of what happens to our data online is still prominent today. Only 59% of people are willing to share information regarding their internet navigation with companies even though they have trust in them and almost all (94%) wish to have the possibility to delete at all times their personal information that business can use. Although highly understandable, companies need information on their target audience to deliver relevant content, so this presents tremendous challenges they have to face.</p><p>These statistics concerning the sharing of information and protection of data go hand in hand with another hot topic these days which is the use of cookies and whether they should be kept or not. Google has already announced that by December 2021, they will completely stop the use of third party cookies and search browsers such as Mozilla Firefox have already stopped using cookies on their platforms. With the protection of data at the forefront of the challenges facing internet and online consumption, media and business will quickly have to find new ways to innovate and adapt.</p><p>One way to approach this challenge for media is using a technology able to understand the context of what a viewer’s interest is at a given time, not through the use of cookies to have access to their search history, but rather by understanding what they are currently viewing. Having access to such contextual information and being able to analyze it will help advertisers select targeted ads in a much safer way which would require much less registered data on users and be more compliant with GDPR rules.</p><p>There is a lot to bet than the reality will probably be halfway, through a clever mix between qualified personal data and contextual data. But what this survey brings to the table is the necessity for media and brands to rely ever less on users to improve their targeting, if they do not want to risk loosing their audience.</p><p><strong>PROVIDING INNOVATIVE WAYS TO TELL STORIES</strong></p><p>How can the media industry couple in-depth storytelling and use of artifical intelligence ? At first sight, these two do not quite match together, as artifical intelligence could be seen as a way to deprive any content from human touch. However, the past years have shown that Artifical Intelligence could in many ways serve stories on three main issues.</p><p>First, by improving interaction with the audience : automating personnalized content pushes, creating innovative formats, creating white lists and black lists depending on the user or the brand’s requirement, giving the ability to navigate throughout a chaptered content are as many examples where artificial intelligence has proven efficient, both in quality and quantity in comparison to human managed processes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1018/1*iG6fnNYA2WIE2RKFBZ2CRQ.png" /><figcaption>Credits — Tomorrows News Report’s from Reuters</figcaption></figure><p>Second, by bringing transmedia storytelling to a new level. Artificial intelligence allows even more to understand context of appearance and resume content or brand’s presence on various platforms and devices, whenever it is stopped. Over time, media consumption has shifted from a very linear and geo-located frame to fragmented and globalized habits. Automating a contextual approach is the only way to learn from these new habits and adapt content to consumption. To make sure content will adapt to the daily worker watching news in the subway as well as the occasional reader who has 10 minutes to kill on a sunday at home.</p><p>Third and not the least important, by finding the right match between content and advertising. 57% of people judge the advertisers they see on a trusted news website can influence their opinion of the news brand. In this sense, a technology having the ability to instantly analyze a content on a video to help advertisers and publishers keep track of who they’re going to be affiliated with in terms of advertisement is crucial.</p><p>No wonder then that 86% of respondents are judging artifical intelligence very likely to influence news consumption in the future.</p><p><strong>KNOWLEDGE BUILDS TRUST</strong></p><p>Throughout this survey, the idea of providing a secured, trusted and personalized context to any information will become more and more essential to the relationship users build with online media. This is even more so important for video, as a single hateful content, fake news video, wrong advertisement or non relevant piece of information, can tarnish the media reputation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/760/1*LWpOmGB9K9iX0wO_vY6qAQ.jpeg" /><figcaption>Crédits : THANANIT — stock.adobe.com</figcaption></figure><p>At Reminiz, it’s been a long time since we know for sure that the only way to build a trustful relationship, with a media, with an audience, with a brand is to gather knowledge on content we provide. <a href="https://medium.com/reminiz-insights/content-data-at-scale-the-underrated-eldorado-for-video-platforms-e01784456940?source=collection_home---4------1-----------------------">The ever growing flow of information and especially video content has made purely impossible to rely on human-based processes</a> to ensure this trust. Artificial intelligence in this case, is not a futuristic and meanless word : it represents the ability to rely on technology in order to create knowledge instantly out of any piece of video and take decisions based on shared and common rules on whether a content is likely to be a problem for the audience.</p><p>The good balance is still to be found to make sure contextual understanding is used at its best to build trust, secure data privacy and provide innovation to media consumers.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=12ece2208b9e" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/trust-data-security-innovation-a-brief-look-at-tomorrows-news-report-12ece2208b9e">TRUST, DATA SECURITY &amp; INNOVATION — A BRIEF LOOK AT TOMORROW&#39;S NEWS REPORT</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Content Monetization at the lockdown era : now’s the time to play it smart]]></title>
            <link>https://medium.com/context-insights/content-monetization-at-the-lockdown-era-nows-the-time-to-play-it-smart-8f1267c223ce?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/8f1267c223ce</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[medium]]></category>
            <category><![CDATA[covid19]]></category>
            <category><![CDATA[advertising]]></category>
            <category><![CDATA[content-marketing]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Tue, 14 Apr 2020 08:54:21 GMT</pubDate>
            <atom:updated>2020-04-14T08:54:21.527Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>Content Monetization at the lockdown era : now’s the time to play it smart</strong></h3><p>For the past weeks, due to Covid-19 pandemic, all industries around the globe have faced a complete shutdown or at least a very significant slow-down. At a time when many countries have organized a full population lockdown, one above all should be delighted: the media industry. How come many advertisers keep lowering their budgets though?</p><p><strong>MEDIA CONSUMPTION AT ITS BEST</strong></p><p>As we are forced to stay home, some without any professional activity, <a href="https://www.comscore.com/Insights/Blog/What-consumers-in-Europe-do-online-during-the-lockdown">media consumption has skyrocketed</a>. Online consumption has increased on all fields, from entertainment to news, video games or search. In Europe, in major countries where lockdown is active, social medias have faced a +30% increase of consumption. In Italy or Spain, information medias have benefited from a +60% increase of traffic. Same observations for OTTs: <a href="https://www.tvtechnology.com/news/netflixs-global-popularity-skyrockets-amid-coronavirus">after the first week of April, Netflix’s popularity score sits at 100, a 142% increase from the previous year.</a></p><p>As for television, first studies have clearly shown that <a href="https://www.comscore.com/Insights/Blog/US-TV-Viewing-Is-Increasing-During-Coronavirus-Pandemic?utm_campaign=EMEA_REG_MAR2020_NRTR_COVID19&amp;utm_medium=email&amp;utm_source=comscore_elq_EMEA_REG_MAR2020_NRTR_COVID19_T3_RU&amp;elqTrackId=019371380c6042bda3a60652825849a1&amp;elq=7fddecd55008406e8b9a45b44112bcc0&amp;elqaid=10379&amp;elqat=1&amp;elqCampaignId=6520">the average consumer spends more time watching TV per day</a>. Cable news networks viewing has increased 73% compared to the same period last year, as the big four broadcast has “only” increased 19%. Finally, some specific channels and programs have the wind in their sails: financial programs and daytime children programs are favoured.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ys9xtY2nBULLrSVF-keSfg.png" /><figcaption>An exemple of digital increase of consumption</figcaption></figure><p>Looking at it, whatever the country, the medium or the hour, all medias have clearly benefited from the lockdown, observing a significant increase of viewing rates and time of consumption per user.</p><p>In the meantime, advertising spendings have actually gone the other direction.</p><p><strong>LOWERING AD BUDGETS: A SHORT TERM STRATEGY</strong></p><p>As <a href="https://www.linkedin.com/pulse/history-advertising-recession-kunal-gupta-1e/">this paper from Kunal Gupta</a> states it very clearly, 24% of brands have paused all ad spend for Q2, while digital ad spend and traditional channels are respectively down 38% and 43%.</p><p>This tendency actually touches all medias and platforms, including social medias. When brands did not cut off the entire budget, they at least delayed major campaigns and decisions, waiting to have more insights on the crisis outcome.</p><p>As for television, all channels have faced advertising decrease, due to brands cancellations or campaigns delays. As a result, advertising tunnels between programs are getting shorter and programs are less remunerative despite higher audiences.</p><p>This global and collective ad-spending decrease has many explanations. This first one is linked to budget cuts, obviously. As companies are starting to feel the heat of the crisis to come, they start paying attention to every penny they can save. As always, communication and advertising are the first ones to sink.</p><p>The second reason can be found in the incapacity for brands to get a direct “feedback” on their campaigns. What is the point of advertising for a new car if I’m not able to sell it within the next two months? Obviously, this short-term reasoning will have serious damages on the long-term. As Gupta explains: <em>“Kantar estimates that brands who go dark to save costs will see a 39% reduction in brand awareness and delay recovery after Coronavirus.”</em> In other words, not advertising now because you can’t sell, will lower your global brand awareness and reduce your sales afterwards. 2008 crisis aftermath regarding advertising strategies has been a <a href="https://www.forbes.com/sites/bradadgate/2019/09/05/when-a-recession-comes-dont-stop-advertising/#6524cebc4608">very clear example of it</a>.</p><p><strong>LACK OF CONTENT: THE MEDIA NIGHTMARE</strong></p><p>A third reason can explain the advertising spend decrease: as video production is almost completely at rest, channels, social platforms &amp; digital medias have less original pieces of content to display, thus less interest for advertisers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*rOCS92Zk27vQaMTRoY-QpQ.jpeg" /><figcaption>What’s next to watch ?</figcaption></figure><p>In a world where the audience expects something new every single minute, such a lockdown is a true nightmare for the video production industry. Television first, was forced to take important decisions: cutting-off their talk shows (no guests anymore), their live shows and splitting their running shows (such as reality competition series) in half to make it last longer.</p><p>In parallel, traditional and digital medias have be overflown with stressful Covid-19 content, whether medical, social or financial. In this context (no new exciting content and repetitive scary information) it is not so surprising that advertisers have decided to cut off their budgets.</p><p>As publishers and channels are used to distribute so much more valuable content than they are currently doing and advertisers are cutting off their spend, what is left to do?</p><p>Could smart reuse of massive content inventories and contextual targeting save the day?</p><p><strong>FOR PUBLISHERS: MONITORING AND MONETIZING YOUR OLD INVENTORIES</strong></p><p>Facing the lack of new productions, publishers have three choices: accept the lack of content on their platforms and the adequate decrease of advertising spending, replace content by 100% Covid-19 content or create new content out of old one.</p><p>This last solution seems to be especially valuable for talk shows or sports shows, blessed with thousands of valuable hours of content that many would love to see again in condensed and adapted formats.</p><p>Creating such formats requires lots of human resources … or a <a href="https://medium.com/reminiz-insights/content-data-at-scale-the-underrated-eldorado-for-video-platforms-e01784456940">good investment in computer vision technology</a> to monitor and monetize massive inventories. Now’s the time to create premium revenue out of reused content. The addition of small bits, video per video, creates overall way higher value than repetitive content on Covid-19.</p><p><strong>FOR ADVERTISERS: A CONTEXTUAL APPROACH TO CREATE A LINK TO YOUR AUDIENCE</strong></p><p>This reuse strategy linked to a contextual approach (finding the right video context for the right audience) gives strong arguments for advertisers to maintain budgets during the lockdown and work on their long-term awareness and relationship with the audience.</p><p>More than ever, <a href="https://medium.com/reminiz-insights/a-year-to-remember-the-true-advent-of-contextual-advertising-c93bc63bd43f">context of appearance will matter</a>. There is no doubt than a brand is less inclined to appear next to a Covid-19 news than a refreshing content… unless it has something to say about Covid-19. At a period when all users are sensitive about the global situation, advertisers and content publishers have a mission to rethink their way to address the audience, to put it in the right context.</p><p>Publishers need to provide more various content in order to offer the right contextual opportunity for advertisers, thus to create a strong relationship with the audience during these dark times. There is no doubt that brands doing so will strengthen their position on the long term.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8f1267c223ce" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/content-monetization-at-the-lockdown-era-nows-the-time-to-play-it-smart-8f1267c223ce">Content Monetization at the lockdown era : now’s the time to play it smart</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What happens when manual moderation ceases to exist ?]]></title>
            <link>https://medium.com/context-insights/what-happens-when-manual-moderation-ceases-to-exist-648ef49d6c91?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/648ef49d6c91</guid>
            <category><![CDATA[moderation]]></category>
            <category><![CDATA[video-marketing]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[social-network]]></category>
            <category><![CDATA[content-marketing]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Wed, 25 Mar 2020 13:30:04 GMT</pubDate>
            <atom:updated>2021-01-19T13:48:49.686Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>What happens when manual moderation ceases to exist ?</strong></h3><p>Since the Coronavirus has stepped into our lives and shut down a huge part of people’s activity, companies have been forced to make strategic decisions to adapt to new restrictions, including the cessation of work of their employees. Facebook, Google, Twitter and others have faced such reality regarding their most invisible employees -or partners- : moderators.</p><p>These front-line workers have a crucial role in keeping social platforms safe from any content that could either harm their image and even worse, risk any legal action against them. What happens then, when manual moderation ceases to exist and that social platforms are “only” left with their Artificial Intelligence to moderate ? A leap of faith for the internet giants.</p><p><strong>Outsourcing the dirty work</strong></p><p>Moderation outsourcing has become a standard over the years, for three main reasons.</p><p>First because no current publisher or social platform has been capable of fully automating moderation to this day and that all of them have been overwhelmed by the volume of content they are supposed to handle every day. We have <a href="https://medium.com/reminiz-insights/towards-a-sustainable-approach-to-save-the-video-ecosystem-2717f066e5bb">already discussed on our blog</a> why the ever-growing quantity of daily content has left the stakeholders with the impossible challenge of keeping manual moderation as a standard. At least, outsourcing represents a short-term solution to reduce the cost.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*00FnWoJqNkmlh6DItyu8Ww.jpeg" /><figcaption>Typical working hours nowadays</figcaption></figure><p>Second because outsourcing to a partner enables gathering all moderators in the same production site with strict control over who gets in and gets out. All this to avoid any data leak that could be very much harmful for the company. Third because the working conditions are amongst the worst in the planet and that many workers <a href="https://www.theverge.com/2020/1/24/21075830/youtube-moderators-ptsd-accenture-statement-lawsuits-mental-health">have suffered from several sorts of mental disorders related to stress</a>, making it one of the jobs with the highest turnover rate.</p><p>In other words, tech giants have outsourced the dirty work to limit the inherent cost in the short term, to protect their data with very strict conditions and avoid having to deal with lawsuits and rising turnovers. The combination of these three reasons explains why the Coronavirus outbreak has made it impossible to keep them working. They would have to either make them come and risk spreading the disease or make them work from home and risk any data leak.</p><p><strong>Time to trust the A.I</strong></p><p>The era we have stepped into is an era of full trust towards the A.I then. Social platforms have no other choice (well, apart from shutting their platforms) than to keep the last moderators in place : artificial intelligence. And this is how it gets tricky. Depending on the length of Coronavirus restrictions and the volume of harmful content published, one could quickly assess the actual quality of each and every stakeholder regarding automatic moderation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1023/1*jtT5-_uD1JKZEXqcvoQnhA.jpeg" /><figcaption><em>Credit : The Washington Post</em></figcaption></figure><p>Social platforms and content publishers use various range of tools to help human’s decision regarding moderation and it is essential to understand that they go from very advanced solutions making pretty much 90% of the job to quick-fixes only designed to ease the human’s tasks. In this second case, one could bet that results without human moderation will lead to epic fails. Text and image solutions can probably achieve an acceptable level of satisfaction. Video solutions, however, are still in the verge of being used by these companies and represent a much higher challenge. Why then ?</p><p><strong>Video makes it tricky</strong></p><p>The main reason is simple: a video is an evolving item, in which each image is linked to another. It is one thing to understand an image. It is another to make assertions about causal links existing between millions of images in the same content. The gap between both is tremendous. And explains why video understanding is not such an easy feat.</p><p>Video understanding can only exist if two conditions are met: automation and causality.</p><p>Causality can be summed up by a simple example: in a video you’re watching on your favorite social platform, two people are kissing. If you take single images of the scene, you’ll think that there’s nothing harmful. They’re kissing, duh. But now if you analyse the whole scene, you’ll notice that before the kiss, the guy is acting roughly with the girl, pushing her a bit, blocking her way and that after the kiss, emotions detected on the woman’s face show fear, disgust and sadness much more than love and happiness.</p><p>Instead of a “simple” kiss, you might have to deal with sexual harassment, letting you to believe that this has nothing romantic and should even be reported. Your perception of the scene has fully evolved. Of course, you did not have the context. Neither did your algorithm.</p><p>That simple example leads us to understand why fully automated video moderation represents a much bigger challenge to those that did not invest (or so few) into such solutions yet. Apology to terrorism might as well be hidden in a “comedy” video. A sex scene might occur in the middle of an apparently safe program. And the presence of weapons could be discovered in the middle of a tutorial.</p><p>Add to this the volume of video content to process and you’ll have the reason why fully trusting their moderation A.I is a leap of faith for the tech giants.</p><p><strong>Time for awareness</strong></p><p>As Coronavirus might lead us to more awareness regarding our fragile human condition and the way we build our future, the lack of human moderation for a while could represent a true epiphany for social platforms and content publishers : human cannot deal with content moderation anymore, for many other reasons that the simple fear of publishing harmful content.</p><p>Here are a few reasons amongst many.</p><p>Having unlabeled and unverified content makes stakeholders lose increasing amounts of money. The quantity of daily uploaded content compared to the inherent cost of its storage, its management, and its publication, rightfully raises the question of this model’s sustainability.</p><p>Videos of lesser quality being so hard to monetize by human means, all stakeholders bet on the same videos which present a high potential. Thus, there’s a total absence of differentiation, leveling of prices from the top, and a waste of 90% of inventories who are not used because they’re not categorized and referenced.</p><p>The online video market is increasingly confronted with the pressure of possible flaws in content management, scandals due to explicit content, respect of Intellectual Property. Content libraries have become larger than the means used to control them.</p><p>A market which has become uncontrollable and stakeholders who lose money gradually as they generate and distribute content: here is the paradox at a time when videos have flooded each fragment of the web. Here is the time for these stakeholders to rely on a single solution to regain control over two core problems regarding their content : safety and monetization. And blindly trust A.I powered moderation.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=648ef49d6c91" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/what-happens-when-manual-moderation-ceases-to-exist-648ef49d6c91">What happens when manual moderation ceases to exist ?</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TOWARDS A SUSTAINABLE APPROACH TO SAVE THE VIDEO ECOSYSTEM]]></title>
            <link>https://medium.com/context-insights/towards-a-sustainable-approach-to-save-the-video-ecosystem-2717f066e5bb?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/2717f066e5bb</guid>
            <category><![CDATA[medium]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[video-marketing]]></category>
            <category><![CDATA[content-marketing]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Tue, 15 Oct 2019 11:59:57 GMT</pubDate>
            <atom:updated>2019-10-15T12:03:14.093Z</atom:updated>
            <content:encoded><![CDATA[<p>More than 300 hours being uploaded on YouTube every minute. The average US citizen having access to 4500 hours of daily live content through its TV provider, 50% more than 10 years ago. Facebook employing 4500 people to moderate video content and about to hire 3000 more. What do these figures have in common ? The ever-growing uncontrollable nature of video production on the internet and the tremendous waste and cost it will represent in a near future.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*v3YBieqDeMXOo7ZCbZZzHQ.jpeg" /></figure><p><strong>POINTLESS, IRRELEVANT AND UNPREDICTABLE</strong></p><p>The evolution of a cloud-based ecosystem and access to costless video production tools have left us thinking that all video production uploaded, stored and played day after day were intangible assets with a zero cost. Huge mistake. Video has become the standard of the media industry, but the outcomes of such infinite production have not been taken seriously. Should and will our salvation require a sustainable video production ecosystem ?</p><p>The truth is media companies still do not really know what to do with video content they owned, own and will own :</p><p>· <strong>Past content is pointless</strong> without any real reuse strategy : imagine a giant library where you dump millions of books every day, with no real label and convince yourself that you will have complete ability to dig into it 2 days, weeks or months later. Thank you, AWS and its huge storage capacities : media industries now store trillions of video content they will never reuse.</p><p>· <strong>Present content is irrelevant the minute it’s online</strong>. Once it’s published, video content is already expired. At least we think this way because publishers and media owners lose almost immediately track of the life and death of a content on the internet. To the best, they keep score of views on their own platform and move to the homepage video a few hours later.</p><p>· <strong>Future content is unpredictable : </strong>it requires too much human resources to moderate upfront and by the time you know what to do with it and if it’s safe for work, you already have lost your launch window.</p><p>How to sum-up video production on the internet in a few words: don’t know what, don’t know where, don’t know when. However, the costs linked to such unmanageable strategy are tremendous.</p><p><strong>LOSS OF PROFITS AND EXTRA COSTS</strong></p><p>Video content accumulation for any media industry represents both loss of profits and additional costs. The loss of profit is linked to the short timeframe available to capitalize on a video Vs the number of videos to process. There used to be a time when a single video content had all your attention to make sure you would target it properly, make sure its content is safe, contact the right ad partners to bid on it and think about the platforms you’d push it on. Multiplying the operation to the infinite day after day has making it impossible to operate quantitative AND qualitative campaigns. Automation started to penetrate all levels of the content distribution process, except one part : assessing properly what was in each video, de facto reducing the value of each video, due to bad targeting, unsafe environments or irrelevant advertising.</p><p>In the meantime, extra costs went through the roof : human moderation by the thousands, increasing storage capabilities and time-limited content became the standard. Less profit, more cost and no concrete solution to lean towards a more sustainable model. Looking closer, the video content industry is pretty much looking like a dump running out of ideas to contain the new waste waves about to come.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*htpkJIC82SvEgOmZqku_Qw.jpeg" /><figcaption>Metaphorically, your video DB might look like this in a near future</figcaption></figure><p><strong>THE CLIMATE CHANGE METAPHORE</strong></p><p>If we do look at produced, published and stored videos like waste, which they are considering the environmental footprint of their data centers, you might face three different scenarios within the years to come, depending on how serious you consider the question :</p><p><strong>The climate sceptics approach</strong></p><p>No real care for the extra costs to come, as long as it makes money now. You keep storing for nothing, keep pushing the industrialized way with the help of an army of human resources for moderation. You bet that when time comes, you’ll find other solutions. At Reminiz, we do think you’re already doomed.</p><p><strong>The local reasoned approach</strong></p><p>Produce less but better, giving the favor to push and store “local” content, that has a real value, that you know of. Mass storage is out of the map and most importantly, each content is judged by its intrinsic value. This approach is probably as beautiful as unrealistic: how would you contain a media that already represents 80% of the internet traffic ?</p><p><strong>The sustainable approach</strong></p><p>Well, no surprise, this is our favorite. Acknowledging that video has now become a central media and will even more so established itself as the unique medium for most people does not mean we can’t do anything about the actual downsides of it.</p><p>What would be a sustainable approach of video content then ? For once, it would mean learning how to “recycle” existing content into new ones by mixing and matching bits and parts of them, through “snackable” content. One would not consider a video as a single entity anymore, only valid during a limited period of time but as a combination of timestamped parts available from your library and easy to replace in another content whenever needed. True reuse.</p><p>For twice, it would mean completely automating video moderation to reduce cost related to pre-publishing and then to storage. Produce once to publish multiple times, publish only safe content, store only parts you will reuse.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/700/1*uX60zxP9emj5u1jjrjFWGQ.jpeg" /><figcaption>A clean and tidy virtual library, but less scary than in the matrix</figcaption></figure><p><strong>IT ALL STARTS WITH UNDERSTANDING</strong></p><p>Nothing sustainable can be achieved without true understanding of content. How would you want to recycle material if you don’t know its components ? This is why Reminiz grew as a company to address all kinds of video understanding and became an expert at facial recognition, logos and brand identification, settings, emotions, and objects detection, making certified assumptions on all kinds of interactions between various elements of a scene, from a gentle kiss to a ship sinking.</p><p>This detailed approach of video understanding enables infinite new possibilities to create sustainable video strategies and operate directly on :</p><p>· <strong>Your brand safety strategy</strong> : Set-up filters to make sure unsafe content is identified and blocked, whether on objective criteria (explicit content, violence, drugs etc.) or on personalized requirements.</p><p>· <strong>Contextual advertising</strong> : Create profiled video inventories fully aligned with your advertising strategies for a contextual approach and an increased likability. Make sure your ad appears on the right videos at the right time.</p><p>· <strong>Live Stream Monitoring</strong> : Monitor streams, channels and any content in real-time and increase live interaction with viewers depending on content and context of appearance.</p><p>· <strong>Content recommendation at scale</strong> : Automate large video volumes processing, replace human-based tags by verified and homogeneous tags at scale with minimum delay to improve instantaneous and continuous content recommendation to users.</p><p>In the end, it all comes down to the ROI you’ll generate for each video you’ll take the decision to host, push or even produce. Through this well-thought model and with the right tools, the return on investment can significantly grow, whereas keeping the “polluter’s strategy” will only increase the cost and lower the value.</p><p>The video industry is at a crossroad and only visionaries adopting a sustainable strategy, from production to storage, will manage to survive the tremendous costs of their ever-growing content libraries.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2717f066e5bb" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/towards-a-sustainable-approach-to-save-the-video-ecosystem-2717f066e5bb">TOWARDS A SUSTAINABLE APPROACH TO SAVE THE VIDEO ECOSYSTEM</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Brand Safety and video understanding : a matter of life and death for the media industry (Part 1)]]></title>
            <link>https://medium.com/context-insights/brand-safety-and-video-understanding-a-matter-of-life-and-death-for-the-media-industry-part-1-57b8a359241a?source=rss-ec3e3a9f2193------2</link>
            <guid isPermaLink="false">https://medium.com/p/57b8a359241a</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[video-marketing]]></category>
            <category><![CDATA[vidéo]]></category>
            <category><![CDATA[content-marketing]]></category>
            <category><![CDATA[brand-safety]]></category>
            <dc:creator><![CDATA[Paul Chaumont]]></dc:creator>
            <pubDate>Mon, 15 Jul 2019 12:32:51 GMT</pubDate>
            <atom:updated>2019-07-16T13:31:09.110Z</atom:updated>
            <content:encoded><![CDATA[<h3>Brand Safety and video understanding : a matter of life and death for the media industry (Part 1)</h3><p>Brand safety has exploded over the past couple of years and has become a vital priority for the media industry. Advertisers cannot afford any single mistake on giving the impression they endorse inappropriate content and content providers are dealing with such inventories it has become mandatory to sort the good from the bad as fast as possible. However, even though automated on semantic understanding, it seems like video Brand Safety still follows very archaic standards, both in categorization and automation. It can easily be explained by the difficulty to assess each item and event occurring in a video.</p><p>Many Brand Safety “experts” thus rely on human platforms dedicated to human moderation with barely satisfying results. Video understanding is without a doubt the ultimate key to help brand safety step into an automated and 100% verified process. Without it, would the media industry be able to survive the tsunami of unverified video content they are already facing and will even more in the future ?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/890/1*2LVJzjfN_v8avS693j7Cjw.png" /></figure><p><strong>Brand Safety 101 : what to put behind ?</strong></p><p>This might be the trickiest question of all. Brand Safety in its broad meaning is a set of measures aiming at protecting a brand’s image from any content that might deteriorate it in an advertising context. In other words, making sure that your brand is not associated to any content that might ruin its effect. Brand Safety has evolved throughout the evolution of web content : not only did it become wider and trickier to delimit but the exponential number of content and its various forms made it an exhausting but vital task for all.</p><p>When speaking about Brand Safety, one usually follows the same broad categories that we all know, such as violence, alcohol, drugs, hate speeches, porn and offensive content. Within time though, advertisers and brands raised their expectations in order to adapt to new forms of “offensive” content : illegal download and most recently fake news are now a priority for the industry, as streaming platforms and social networks took over the internet.</p><p>Investing in Brand Safety should not be reduced to protecting a brand with any negative impact : it actually has much broader consequences especially if taken lightly.</p><p><strong>One mistake to ruin a whole campaign</strong></p><p>Always remember it takes one mistake to ruin a brand image on the long term. It has been proved that consumers always viewed advertising on a content as intentional from the brand, whatever the context. If a brand were to appear on a YouTube channel with racist undertones, consumer would most likely retain that the brand endorsed the content, therefore is a bit of racist itself.</p><p>Though hard to measure, <a href="https://www.cheq.ai/the-brand-safety-effect">a recent study</a> has tried to quantify the direct viewer effect of placing an ad next to an offensive content. This study indicates a 200% brand purchase intent decline after watching inappropriate content as the brand quality perception is 7 times worst. Meanwhile, consumers are 450% more likely to feel the brand does not care about them.</p><p>Brand Safety fails can have catastrophic financial consequences as they truly have an impact on the brands perception and future purchases.</p><p><strong>Human moderation : a costly and exhausting mission</strong></p><p>Brand Safety process still involves a lot of human tasks. Although lots of specialized companies claim they have performing algorithms to moderate any offensive content, they usually rely on outsourced platforms of hundreds of human eyes binge-watching images and videos and sorting the good from the bad ones. Very far from full automation then.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/759/1*l-CjHOoq_xd1e5su98XSZA.png" /></figure><p>But all this has a real cost. First, because humans cannot watch more content than they have hours in their daywork. The more content, the more resources needed to monitor it. Second because humans make mistakes: if you want to feel what it’s like to take a decision on a piece of content after having watched eight straight hours of it, you’d be surprised how you end up losing any rationality. Third because of cultural bias: many moderation platforms are outsourced and depend on ever-changing cultural references. Again, a human eye will always have biases preventing him to operate full objective moderation, even with clear guidelines.</p><p>Last but not least, because it is the worst job in the planet. And how could this be different : imagine spending hours watching violent content, porn or incitement to terrorism all day long, every day. In this kind of business, turnover is very high, and employees usually end up losing it pretty quickly.</p><p><strong>Preventing rather than reporting</strong></p><p>As companies still rely on human based moderation, even though they like to hide behind obscure technological solutions, they usually don’t have the capacity to prevent inappropriate content from being published. Every minute, 400 hours of content are uploaded only on YouTube. Assuming that you’d want to do a proper job and ensure full brand safety, this lets you imagine how many human platforms would be required to handle this. By the way, this is exactly the kind of problems companies like Facebook are facing right now, trying to avoid any inappropriate content to remain on their platforms longer than minutes. As preventing would create an infinite waiting line and eventually kill the magic of instant publishing, the only solution remains reporting and taking the risk to push inappropriate content out there, waiting for any advertiser to fall in the trap.</p><p>Brand Safety automation and adapted algorithm to detect any irregularity would not only secure the whole process but ensure never publishing content at risk.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/650/1*Xml3misOOZ2lCzSwL-BhAA.jpeg" /></figure><p><strong>Towards modular Brand Safety…</strong></p><p>As long as Brand Safety will remain a matter of human-based activity, it will never get the chance to be as accurate as brands will like it to be. Sure thing, any brand is capable of telling you they do not want to appear next to gun shootings, ISIS flags or Nazi salutes. But Brand Safety is so much broader than this and more importantly, adapted to each brand. Football star Neymar sponsors having trouble benefitting from his current image would probably want to avoid strengthening this endorsement on videos featuring Neymar’s extra-sportive issues. At least for a while. To the contrary, if you’re a non-profit fighting against animal cruelty, you might have interest in appearing in these exact video content featuring animal cruelty to make your point. In other words, Brand Safety is as polymorphic as there are companies and communication objectives.</p><p>This is why video understanding allows the media industry to lean towards a modular Brand Safety. Depending on your brand, it is the promise to activate or deactivate any criteria that might harm your brand image.</p><p><strong>…And inventory optimization</strong></p><p>Finally, it is not to be forgotten the power of video understanding when it comes to automate sorting of thousands of videos to be published everyday and offered as advertising territories to the brands. The quicker and more accurate a publisher can be to sort good advertising material from bad material for a client, the higher the chances to monetize a larger spectrum of the inventory on a longer time.</p><p>Indeed, one last thing to take into account is ever changing context. For instance, Harvey Weinstein was perfectly ‘Brand Safe” 18 months ago and any correlation between any advertising and him would not have been a problem. Only Brand Safety automation can allow evolving your standards on what is Brand Safe in the present context.</p><p><strong>Deepfake : the end of human-based monitoring</strong></p><p>Technology has already evolved to the point human monitoring, whatever the amount of money you put into it, won’t be enough to prevent inappropriate content. Fake news have opened new dangerous doors in terms of content safety. The latest one on the subject is deepfake: a technique for human image synthesis based on artificial intelligence, used to produce or alter video content so that it presents something that didn’t occur. Replacing a face by another, adding an item in a video, making people say something they did not say, Deep Fake creates infinite new ways to “play” with content and mess with the integrity of each piece.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*m72tHMhuafR3UWJT3wOyQg.jpeg" /></figure><p>Without a single doubt, Deepfake will soon enough become a Brand Safety priority for any media company. This of course, if video understanding based on true artificial intelligence has started replacing armies of human eyes checking thousands of hours of content.</p><p>In the second part of this article, we’ll tell you more about Reminiz Brand Safety algorithm and how it has been created to operate accordingly to every company’s requirement.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=57b8a359241a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/context-insights/brand-safety-and-video-understanding-a-matter-of-life-and-death-for-the-media-industry-part-1-57b8a359241a">Brand Safety and video understanding : a matter of life and death for the media industry (Part 1)</a> was originally published in <a href="https://medium.com/context-insights">Context Insights</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>