unkz's recent activity
-
Comment on 'The report's so stupid': The DNC 2024 autopsy is roiling Democrats in ~society
-
Comment on 'The report's so stupid': The DNC 2024 autopsy is roiling Democrats in ~society
unkz Link ParentThis seems like good advice though. America is not “on the left”. According to Gallup’s latest figures (2025 annual averages), more Americans described their views as “very conservative” or...This seems like good advice though. America is not “on the left”. According to Gallup’s latest figures (2025 annual averages), more Americans described their views as “very conservative” or “conservative” (35%) than as “very liberal” or “liberal” (28%), with 33% identifying as “moderate.” If you want to be elected, the centre is skewed conservative.
-
Comment on An OpenAI model has disproven a central conjecture in discrete geometry in ~science
unkz LinkThis really echoes what we are seeing with Mythos and exploit generation. LLMs are happy to grind away on the most obscure corners of a problem space for as long as you allow them more tokens,...I believe if the level and type of human expertise that is represented on this note had been assembled to find a counterexample to this conjecture a month ago, and those people put in similar amounts of time working on it tha[t] they did to reading and thinking about Chat GPT’s solution, the mathematicians would have found a counterexample. However, without the claimed proof by Chat GPT, there is no particular reason anyone would have tried to look for a counterexample, assembled a group of experts with the appropriate expertise, or that the experts would have agreed to turn their attention to this problem.
This really echoes what we are seeing with Mythos and exploit generation. LLMs are happy to grind away on the most obscure corners of a problem space for as long as you allow them more tokens, where humans would simply not be interested, and they can do so at a level that is on par with extremely qualified humans.
-
Comment on A random sci-fi question for you in ~talk
unkz LinkNew Mongolia. Let’s say I go to Mars — what do I do when I get back? Probably post on tildes about it, and then here we go again….New Mongolia. Let’s say I go to Mars — what do I do when I get back? Probably post on tildes about it, and then here we go again….
-
Comment on Tesla’s newest electric vehicle could jolt the trucking industry in ~transport
unkz Link ParentI don’t see that as a bad thing though? It seems increasingly likely to me that self-driving long haul trucks are going to end up dramatically safer than humans....I don’t see that as a bad thing though? It seems increasingly likely to me that self-driving long haul trucks are going to end up dramatically safer than humans.
https://pubmed.ncbi.nlm.nih.gov/39485678/
Results: When considering all locations together, the any injury reported crashed vehicle rate was 0.6 incidents per million miles (IPMM) for the ADS vs. 2.80 IPMM for the human benchmark, an 80% reduction or a human crash rate that is 5 times higher than the ADS rate.
That’s mostly city driving as I understand it, which I believe is harder than highway driving so we should see similar or better results. Computers aren’t going to be affected by the main issues truckers face: being exhausted, facing time pressure, and irregular scheduling.
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz (edited )Link ParentReally? You think 4.1 is a competent coding agent compared to 4.5 or 4.6? In my experience at least, it couldn't do even simple tasks without requiring a ton of manual review and fixing.Really? You think 4.1 is a competent coding agent compared to 4.5 or 4.6? In my experience at least, it couldn't do even simple tasks without requiring a ton of manual review and fixing.
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz Link ParentOpus 4.5 was Nov 24, 2025. Opus 4.6 was released Feb 5, 2026. In my experience, 4.5 was the first model that was truly useful for writing large pieces of code, so that’s 6 months. 4.6 was a...Opus 4.5 was Nov 24, 2025. Opus 4.6 was released Feb 5, 2026. In my experience, 4.5 was the first model that was truly useful for writing large pieces of code, so that’s 6 months. 4.6 was a massive upgrade for me when I got access to the 1 million token context, which has been 3 months. I don’t have a very strong opinion on 4.7 — it seems also very good but I’m unsure if it’s materially better than 4.6.
Is 6 months quite a while?
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz Link ParentHow would you define “quite a while”? The way I see it, it’s only been a few months since models became capable of producing high quality code.How would you define “quite a while”? The way I see it, it’s only been a few months since models became capable of producing high quality code.
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz Link ParentThat’s basically what I expect to happen, and what I’m expecting from Anthropic. After all, they don’t have a monopoly on frontier models. Pretty soon OpenAI (and Deepseek, Grok, Meta, etc) will...If it really is that good, we'll see an initial one-off surge and then back to business as usual inside a year.
That’s basically what I expect to happen, and what I’m expecting from Anthropic. After all, they don’t have a monopoly on frontier models. Pretty soon OpenAI (and Deepseek, Grok, Meta, etc) will publish similar results and we will get pre-commit scanning on GitHub so bugs almost never land in a release. Then Mythos won’t be much danger and it will be open to the public.
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz Link ParentHmm, didn’t I address that? I thought I was pretty clear about why I think Mythos is a pretty dangerous tool to give the public right now. We are in a phase of massive bug discovery — once the...Hmm, didn’t I address that? I thought I was pretty clear about why I think Mythos is a pretty dangerous tool to give the public right now. We are in a phase of massive bug discovery — once the backlog has been cleared and we have integrated LLM based security scanning into most of our critical pipelines it will be less of a concern.
Also, look how badly LLMs have wrecked the internet with automated disinformation campaigns just like they predicted. OpenAI wasn’t really wrong about that.
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz Link ParentBut it doesn't really care. If you go tell your team of security researchers that you want them to go an enumerate every possible edge case of a floating point operation in a piece of code that...But it doesn't really care. If you go tell your team of security researchers that you want them to go an enumerate every possible edge case of a floating point operation in a piece of code that has no obvious security relevance, they might just refuse. Like, these people have careers and interests that they want to actually improve instead of wasting their time. There's a kind of publish or perish incentive here where they want to efficiently produce exploits. Mythos will just go do it, and it'll keep doing it for as long as you have GPUs.
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz Link ParentThat’s kind of the interesting thing with Mythos though, right? There are huge classes of bugs that are usually innocuous and so rarely lead to an exploit that they aren’t worth the time for a...That’s kind of the interesting thing with Mythos though, right? There are huge classes of bugs that are usually innocuous and so rarely lead to an exploit that they aren’t worth the time for a security researcher to investigate every possible edge case. Mythos doesn’t care, it will happily enumerate every single variant and chase them down to their conclusion.
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz (edited )Link ParentI imagine that major players like Microsoft are eventually going to be proactively scanning all public commits to any projects that are in widespread use just for self preservation. Far fewer bugs...I imagine that major players like Microsoft are eventually going to be proactively scanning all public commits to any projects that are in widespread use just for self preservation. Far fewer bugs of known classes will even make it into a release.
Anthropic will no doubt dedicate $X million per year for open source scanning. You can see how they are positioning themselves to be a defacto requirement for all corporate internal code commits though. I wouldn’t be surprised to find cybercrime insurance policies starting to put Mythos contracts as a requirement for getting said insurance.
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz Link ParentYeah, anthropic granted participants $100 million in tokens.Yeah, anthropic granted participants $100 million in tokens.
-
Comment on Project Glasswing: what Mythos showed us in ~comp
unkz (edited )LinkThis seems very consistent with what everyone else is saying, and I think it reinforces what Anthropic has been saying about the danger level. The major skill is building the complete working...This seems very consistent with what everyone else is saying, and I think it reinforces what Anthropic has been saying about the danger level. The major skill is building the complete working exploits.
The chief complaint I’ve been seeing from people is that it is largely on par with other frontier models in terms of finding new bugs, and a competent security researcher isn’t going to benefit significantly from Mythos. This misses the point entirely. The danger is any even remotely competent script kiddie can take Mythos and go directly to exploiting live systems, without having to be a competent security researcher.
The other big danger is from advanced persistent threats, who are targeting a specific site. Now whenever a new patch lands, any applicable exploits are gonna be live in 15 minutes. Let’s say I’ve mapped out the attack surface on my target — I know they run Linux, nginx, wagtail, postfix, for instance.
Major threats are already monitoring commits for major things like Linux to reverse engineer exploits, but now this can be done at scale automatically for every stupid NPM or pip package that is included into the requirements.txt or package.json. You get a bug in some obscure xml library and you cue your exploit harness with the knowledge that a commit just landed in beautiful soup and they run wagtail, and you might get an exploit custom tailored to your victim in 15 minutes.
-
Comment on The old world of tech is dying and the new cannot be born in ~tech
unkz Link ParentI wouldn’t say I was willing but rather the story seemed so explicit that I couldn’t see it any other way. After reading the various other interpretations from the first article, I can kind of...I wouldn’t say I was willing but rather the story seemed so explicit that I couldn’t see it any other way. After reading the various other interpretations from the first article, I can kind of picture some of them but they seem very forced (as I’m sure most of those people would say about my interpretation, since my view wasn’t even mentioned and most people are in a totally different theme).
-
Comment on The old world of tech is dying and the new cannot be born in ~tech
unkz (edited )Link ParentAlmost everybody interprets the shapes as being people. This is an interesting article on the video that talks about all the various interpretations that people ascribe to the shapes....Almost everybody interprets the shapes as being people. This is an interesting article on the video that talks about all the various interpretations that people ascribe to the shapes.
https://www.bps.org.uk/psychologist/our-storytelling-nature
And this article focuses more on the autism interaction and why some autistic people take it differently, often more literally. Of course autism is a spectrum, and I’m sure some autistic people will perceive a story and some non-autistic people will fail to.
https://musingsofanaspie.com/2014/04/08/interpreting-the-heider-simmel-animation/
Personally, my interpretation of the film was
It’s an abusive father (big triangle) who is beating his wife (small triangle), then goes after the child (circle), the mother tricks the father by trapping him in the house and they both escape. Then the father has a fit and smashes their house.
I’d be interested to hear other interpretations!
-
Comment on Bun has been rewritten in Rust in ~comp
unkz Link ParentCan’t they just have been wrong about zig? I always thought zig was a stupid choice.Can’t they just have been wrong about zig? I always thought zig was a stupid choice.
-
Comment on Framework reveals 13 Pro laptop with 20-hour battery in ~tech
unkz Link ParentIs this what you are talking about for fixing sleep? If so, yes.Is this what you are talking about for fixing sleep? If so, yes.
Both 1st Gen and 2nd Gen Numpad and RGB Macropad modules share the same hardware and artwork. The 2nd Gen modules feature updated firmware, which includes a fix to prevent the system from waking while being carried in a bag. 1st Gen Numpad and RGB Macropad users will be able to update their firmware in the future for compatibility and to gain the same functionality.
-
Comment on The old world of tech is dying and the new cannot be born in ~tech
unkz LinkIf anyone is interested in watching this video, I assume it’s this one https://youtu.be/VTNmLt7QX8E?si=XnDd1MUKFh-pvAOeSitting in on a talk on autism diagnoses, one of a series of scientific talks, watching an animation they used as a diagnostic aid, hearing everybody around me laugh as if the shapes on the screen made sense, only then truly understanding myself, and feeling more alone than I have ever felt before or since.
If anyone is interested in watching this video, I assume it’s this one
This is not unique to democrats. Reasonable republicans have been derided as RINOs since the 90s, and cuckservatives since maybe 2015.
I actually think part of Trump’s success was a direct result of that “course correction” so maybe there’s a lesson there. Perhaps abandoning the centre and rallying the dirtbag left is a reciprocal route to victory. However, I don’t really think so: my read of the American public there really is a structural disadvantage right now that favours republicans because Americans really are pretty conservative.