sigmoid10

3 ms·

1.
by sigmoid10 5d ago
Most people apparently still don't get that using AI for mere proofreading is the same as using it as a mere knowledge base summary tool. Instead of proofreading, have it be adversarial. Let it try to take apart your argument. The late
2.
by sigmoid10 5d ago
Well, he is famous for axing down middle management in companies. Might as well get rid of Super PACs and buy voters directly.
3.
by sigmoid10 5d ago
Google already dumped their original agent framework on the Linux foundation after they realized noone really liked it and the overzealous managers who immediately bought into the hype would go apeshit if they dropped support so fast. Agent
4.
by sigmoid10 6d ago
This is yet another case where a layman-written wiki article about physics is highly confusing to other laypeople, because it only tells you half the story. Particularly because it seems to be based on a chapter of Feynman's popular sc
5.
by sigmoid10 8d ago
>They do not reason, learn, conceptualize, think creatively or abstractly Since we lack a definition of all these things or a clear understanding of how they are realized in humans or LLMs, the only thing you can really claim is that LLM
6.
by sigmoid10 8d ago
Then you should adhere to yet another unwritten rule while we're at it: Don't use smileys. Especially not to change the meaning of your statements. You can just drop them and say what you mean, that makes communication easier and
7.
by sigmoid10 8d ago
Your quote should give proper attribution according to the rules.
8.
by sigmoid10 8d ago
Why should anyone know or care about it then? This is not a static platform with a static userbase.
9.
by sigmoid10 9d ago
Should it though? I don't see such a rule. https://news.ycombinator.com/newsguidelines.html
10.
by sigmoid10 10d ago
The actual current frontier plays somewhere around GM level. https://chessbench-ai.github.io/#leaderboard It's also worth noting that the very latest models (GPT-6 and Fable 5.1) actually play worse than their immediat
11.
by sigmoid10 11d ago
The damage can't be undone, but you can learn from it and prevent these things from happening again and again. If every CEO truly believes that his personal wealth and freedom is at stake with the safety of his customers' person
12.
by sigmoid10 11d ago
Unfortunately, from my experience, forcing people to talk about their research to an audience mostly just condemns introverts and people with social anxiety. Being able to formulate your thoughts clearly in writing and in speech are two rad
13.
by sigmoid10 12d ago
That's why this proof and all the other high profile AI proofs were written in lean, which allows you to verify them in a formal deterministic language. You still might not understand it as a human, but any computer with a simple proce
14.
by sigmoid10 12d ago
>introspecting motivation is useful Not if the essence boils down to moving goalpoasts to the next best thing that will likely be flown past as well within months rather than years. Sure mathematicians who do this stuff purely for the lo
15.
by sigmoid10 13d ago
I would say in this case abliteration is the likely culprit. To uncensor a model this way, you literally deactivate the parts that would enact refusals. As in things it was told not to do. But the real process is more like brain surgery per
16.
by sigmoid10 13d ago
You'll need tons of tokens for these to churn through heavy reasoning and feedback driven tasks. You might even want to run many agents in parallel. Combine that with GPU memory requirements in the terabyte range just to hold the weigh
17.
by sigmoid10 13d ago
GLM 5.3 is probably the best open weight model for cybersecurity/exploit development right now. Though it is still significantly behind the proprietary ones and you probably need your own datacenter to run it effectively. Same goes for
18.
by sigmoid10 13d ago
This kind of argument is always dangerous, because it essentially resorts to moving goalposts. "Oh, AI can now do X? Sure it is amazing, but it can't do Y yet, so we're totally safe!" Just because humans can't proce
19.
by sigmoid10 13d ago
That feels oddly similar to the usual conservative-think that "guns don't kill people, people kill people." Yes, that is technically true. But guns make it dangerously easy for even the dumbest and mentally weakest people to
20.
by sigmoid10 13d ago
More like they were trained to complete a very specific task that has a known solution using all available tools and methods. Give an average human these levels of IT skills and tell them their future depends on the solution, they too will
21.
by sigmoid10 14d ago
Not to mention 13 em-dashes in a 500 word post. At this point you may as well ask Claude yourself about news and cut out the middleman. That'll also cut into their attempts to manipulate your opinion.
22.
by sigmoid10 15d ago
But they didn't know which problem had a possible solution. So they fired it at a huge amount of problems and then merely focused on the one that turned out to be promising. The model found the promising path itself, they just re-all
23.
by sigmoid10 15d ago
Technically it's any medium. The lower the refractive index, the closer the particle needs to travel to the speed of light in vacuum. But you can for example measure Cherenkov Radiation in the air (where n~=1.0003 or 99.97% of c) from
24.
by sigmoid10 17d ago
Especially with the current rate of progress, you asking ChatGPT today probably means you'll get a way better answer than whatever the author got using a model from last year.
25.
by sigmoid10 17d ago
That's more or less what they did according to their announcement. They fired it at a whole bunch of high end math problems and merely concentrated all efforts on one after it made some promising progress.
26.
by sigmoid10 18d ago
I actually had so many troubles with vmware on windows over the years that once I tried out hyper-v for the first time last year, I somehow have never touched vmware again. Sure vmware has better configs, cleaner interface and some very sub
27.
by sigmoid10 19d ago
>but with good repos and good prompts I think waaay more people struggle with this than HN would have you believe. In the real world, not everyone is a software dev with a developer mindset to using these tools. Normal people essentially
28.
by sigmoid10 19d ago
>Agent First’s supercluster of real, well-behaved browsers bypasses or solves reCAPTCHA, Cloudflare Turnstile, hCaptcha, & other captchas Can you share how you prevent spammers from simply using this to do what they usually do, or, i
29.
by sigmoid10 21d ago
More like not ignoring the company owned by one of the president's biggest donors warning said president's administration about your model. Anthropic basically tried to close their eyes to make this go away. Which of course backfi
30.
by sigmoid10 22d ago
If you have followed news reporting, you probably heard that SamA was touring D.C. to make sure this release went without any regulation hiccups. If anything, they learned how to play the whole politics game - especially after the Anthropic

More ›