Ever since Lion King came out the “wildebeest” name has been much more common.

But the most commonly used logo for the GNU software is an anthropomorphized gnu.

In the software, the official position is that the G in GNU is pronounced. But for the animal known as a gnu, the g isn’t pronounced.

briefly released millions of tracks that were scraped from Spotify via BitTorrent.

That’s just an awkward sentence construction but it makes sense: they released track via Bittorrent. The tracks were scraped from Spotify.

I sold my car that was purchased from a dealership via private party sale.

I charged my laptop that normally accepts 100W via a 20W phone charger.

I would’ve used a “which” phrase with commas to avoid the confusion, but the sentence as written is valid and makes sense.

You can reason from a few principles:

• At its core, the math functions being optimized by these AI tools and their specialized hardware is that they can perform inference and pattern recognition at huge scales across enormous data sets.
• Inferring a rule set for pattern also allows generation of new data that fits that pattern.
• Some portion of human cognitive work falls within the general framework of finding patterns or finding new data that fits an old pattern.

So when people start making claims about things with clear, objective definitions (a win condition in chess, the fastest route to take through a maze, a highest lossless compression algorithm for real world text), it’s reasonable to believe that the current AI infrastructure can lead to breakthroughs on that front. So image recognition, voice recognition, and things like that were largely solved a decade ago. Text generation with clear and simple definitions of good or bad (simple summaries, basic code that accomplishes a clearly defined goal) is what LLMs have been doing well.

On things that have much more fuzzy or even internally inconsistent definitions, the AI world gets much more controversial.

But I happen to believe that finding and exploiting bugs or security vulnerabilities falls more into the well defined problem with well defined successes and failures. So I take it seriously when people claim that AI tools are helpful for developing certain exploits.

but isn’t the memory on the Neo on the same die as the processor?

Not actually on the same die, but in the same package, stacked on top using TSMC’s Integrated Fan-Out Package on Package (InFO-PoP).

So the memory still needs to be sourced from memory manufacturers, sent to TSMC, and then have TSMC package it all together in a single package. It’s unclear whether they had locked up this supply at pre-AI prices, though. The underlying A18 Pro chip/package was annoinced and launched about 18 months ago, so if they had the manufacturing pipeline set up for that they might have kept the contractual rights to continue buying memory at the old prices.

This reporting says that the subpoena requires that Reddit produce the information and appear for a hearing.

Shouldn’t the system be storing timestamps in UTC anyway, and then displaying them in whatever localization settings you have?

On the flip side, I’m a former sysadmin and I only stick around for 5 years because I had the educational credentials to move onto another field (and then another field). I’m glad I did the IT thing in my 20’s, and still like to tinker with homelab stuff 20 years later, but in the end it was a stepping stone towards something else (that does require formal schooling). The degree is a tool that can be used to control on a few more things in your life, in the hopes that you can go where you want to end up.

No, it’s not volunteering, at least not anymore.

Subpoena is legal Latin for “under penalty,” because noncompliance with a subpoena carries a penalty.

Originally, it was an information request from the feds, and Reddit refused. Then they escalated to getting a grand jury subpoena (which means they got a bunch of normal citizens to agree that the information was relevant to a criminal investigation), so now noncompliance carries a penalty.

Reddit notified the users, who hired their own lawyers, who are resisting the subpoena and will litigate it to where they need a judge to decide whether Reddit will have to turn the information over.

That’s the process for these things, and we’re a couple steps in already.

Doordash puts the tip in the order so that drivers can view them as bids for service, and they generally snatch up the highest tipping orders immediately. Low/no tip orders with long distances or a lower rated customer might languish, unclaimed, because drivers won’t want to take them for insufficient pay.

The website could know whether the username actually exists on the system. But revealing that information is a security weakness because someone could at least learn who has an account at that site (especially if usernames are email addresses, as they often are).

The article describes how they immediately went to look for an unsigned 32-bit millisecond counter when they noticed it was happening around 50 days since last reboot, because they already knew that association you describe.

Interesting writeup. Fun little story about the detective work involved.

According to your POV here, companies can claim whatever and it’s my job now to figure out if they are lying or to what extent.

No, the actual claims here, that describe specific bugs in specific software, can be evaluated. Even without whipping out a test environment to try to reproduce the results with your own proof of concept, you can read the text and evaluate whether the claims make sense on their face.

a broken clock is never right, reality momentarily aligns with it, which is a completely different thing

And that’s why the substance of a statement matters. I don’t believe in the supernatural, so if someone says “I’m a psychic and the missing girl on the news is in a shed near the water,” that doesn’t register with me at all. But if that person says “I’m a psychic and the missing girl is in a shed at 1234 Main Street” that raises eyebrows because it is easily falsifiable. And if the person says “I’m a psychic and the missing girl is in a shed, so I looked and found her and reported it to the cops, and here’s a cryptographic hash of my description of how I found her, which I’ll publish once the cops confirm she’s safe” that’s gonna be a much more serious statement. Even if I don’t believe that the person actually is a psychic, I can pay attention to how the whole thing played out because the person claims serious non-psychic validation of the results, and the results themselves are important entirely externally from the claim of whether psychics have powers.

This is a story about several cybersecurity vulnerabilities, some of which sound medium or high severity in very commonly used software. That’s important in itself, outside of AI mattering at all. And if they claim to have the receipts in a falsifiable way, that’s the kind of thing that shows a high degree of confidence in the genuineness of what was found.

I don’t give a shit about AI and I’m generally a skeptic of the future of any of these AI companies. But if someone uses AI tools to discover something new in the subjects that I do care about, like cybersecurity, then I’ll pay attention to the results and what they publish in that field.

This is really a corporate problem of their own making and their responsibility to fix. They have lied so much, I do not owe then a single iota of trust.

The statements can stand for themselves, evaluated on the merits of the claims, regardless of authorship. That’s how these things should work. Someone who has a great history of finding vulnerabilities still has to stand by each exploit/proof of concept they write, on its own merits. On the flip side, the corollary to the adage that a broken clock is still right twice a day is that you can’t just say “oh the broken clock said this so I can ignore it.”

Do you really think any of them would post something like “yeah, we found a vulnerability but it’s basically a typo that could not be seriously exploited”?

The blog post literally describes exactly that, for ffmpeg. And several of the other described vulnerabilities sound like they’re in that category of “here’s a bug but we didn’t find an exploit.”

Simply refusing to engage with these big claims just because of the source is an irresponsible way to approach cybersecurity.

even if the whole scenario is real, it may not have the intervention of Ai they are claiming

…who cares? If it’s a real bug and a real PR addressing the bug, why does authorship or methodology matter?

It’s just the ad hominem fallacy (or the close relative, appeal to authority). Let the actual substance stand and fall on its merits. Read the described vulnerabilities and exploits and decide whether you think those need to be patched and how critical/severe the bugs/vulnerabilities are.

And maybe your priorities are different from mine, but the core of the claim (we found some vulnerabilities) trigger a responsibility to address them (confirm and patch). I don’t care about marketing or corporate interests or whatever in those circumstances, I’m just focused on fixing problems that have been found.

Yes I understand, but I’m also putting the direct claims right there, not filtered through Anthropic’s PR or an article from the IT industry press interpreting those PR statements.

These are real CVEs that have actually been submitted to the code maintainers for both FOSS and closed source software that is foundational to the computing world. Some of them are published in this post. And many more are simply described with a hash of the full writeup indicating that they have it written out and are waiting for the patches to be applied. I’m especially interested in the Virtual Machine Monitor and the exploits for jumping out of browser sandboxes for “all major browsers.”

Some of the published CVEs in the blog post seem pretty serious, especially the FreeBSD remote root access for devices running NFS. The OpenBSD one is a critical DOS vector, and the FFMPEG one is just a bug that doesn’t seem to actually expose the software to any practical exploits but should still be patched.

But they’ve staked it out with their public disclosure of the hashes and a description of a few of the problems. These are big bold claims that are provided in a format that will be easily falsifiable in due time. And treating it as just marketing fluff ignores the shades of gray that actually apply to corporate claims.

The security blog writeup is here:

https://red.anthropic.com/2026/mythos-preview/

They’ve described several patched CVEs and disclosed the hashes of writeups that are currently undergoing the responsible disclosure process.

It lists quite a few, so I’ll be checking back when the vulnerabilities/exploits are patched and disclosed.

We should think about ruggedness and reliability as two different metrics. Some models are quite good at dealing with physical impacts, vibrations, moisture, dust, etc. That doesn’t always translate to reliability for ordinary use, or failures caused by other things related to design or manufacturing, or even software/firmware issues.

LLM companies have argued they should get to ignore all copyright, and now that one of their code leaked, suddenly they care greatly about copyright.

Anthropic itself has argued that digitizing and using the digitized copies to train models is fair use, so long as:

• They don’t redistribute the physical copies they bought
• They don’t allow an end user to retrieve the contents of any one specific work at the user interface (if you ask Claude to spit out the entire text of a copyrighted work used to train it, it is designed to resist copying too much out of a single work)

So they don’t argue that copyright doesn’t count, exactly. They argue that copyright doesn’t prevent model training from ingesting an entire copyrighted work, as long as it’s done with so many other copyrighted works that any given original isn’t a huge contributor to the model or its outputs.

There’s tension in their positions, but not so much that it would totally fall apart.

God I wish democracy meant that we could vote on decisions like this

You can! Only problem is that it’s one vote per dollar instead of one vote per person.

It sucks though, I agree - software should get more efficient over time, just like hardware does.

It generally does, for any given computing task, but the problem is that generally software adds more features over time, not least of which is supporting new hardware that hits the ecosystem.