daveguy

5 ms·

31.
by daveguy 24d ago
But also, fuck trump.
32.
by daveguy 24d ago
Getting the kids to insist they be paid what their contributions are worth would raise the boat for all developers. Of course that would take some sort of big bad union, which the oligarch vc asshats really don't like.
33.
by daveguy 25d ago
The coming shitstorm of crappy software is going to suck.
34.
by daveguy 25d ago
Nah, that just makes it a shitty tool.
35.
by daveguy 26d ago
Slick bullshit. That's AI.
36.
by daveguy 26d ago
I don't think they meant the nurse would be responsible. The LLC would and should be fucked in that case.
37.
by daveguy 27d ago
The question is more whether OpenAI can be trusted to honor that toggle switch. Given that they hold all chats for 30 days "for safety and security".
38.
by daveguy 27d ago
#2 (prove those chats changed model behavior) is pretty straightforward if the anonymized data from chats can be actively searched by a model. In fact, it could be very clear if the provenance of context is traced. If anonymized data from
39.
by daveguy 27d ago
If the model has access to the "anonymized" data from chats, and the model is capable of building its own context from data that it can search through, including this data. Then it looks pretty damning. An independent review of
40.
by daveguy 29d ago
But you get more funding when you call it Recursive Self Improvement. Even better if you call it RSI so it doesn't evoke pesky skynet scenarios outside of AI safety circles.
41.
by daveguy 29d ago
This only applies if LLMs aren't making mistakes 20% of the time and that's the problem. When you're only saving time on the easy part, it doesn't matter if you're working twice as fast because review of the tricky
42.
by daveguy 29d ago
Nice strawman. I doubt the white supremacists / nationalists / nazis even want you to come to their defense if you're going to be weak about it.
43.
by daveguy 29d ago
But understanding that requires reading the code or writing that it spits out rather than having yet another LLM do it for you.
44.
by daveguy 29d ago
You're right that Nazi doesn't mean Nazi anymore. It means neonazi / white supremacist / white nationalist, which is a much broader group of people that, for some baffling reason, are under the impression that people do
45.
by daveguy 1mo ago
Only problem is the LLM comments are the exact same quality as the code. In other words, LLM comments also have to be thoroughly reviewed.
46.
by daveguy 1mo ago
The nasa guides are a great resource. Designed with the knowledge that a bad soldering job can cost lives. Maybe overkill for your average diy, but great for a thorough high-quality deep dive. Here is the reference standard context of that
47.
by daveguy 1mo ago
Crypto mining would be one application. But also, combined with a sandbox escape would make it particularly devastating. Usually full control of a device takes at least two exploits given the layers of security present in OS and browser env
48.
by daveguy 1mo ago
Not entirely. Lack of regulations mean recycled water facilities can end up being consumption facilities with pollution risk. Also Meta plans to restore as much water as they use by 2030. Which, seems like a while to break even. That's
49.
by daveguy 1mo ago
Exactly right. It's a lot easier to prevent an asshat venture than it is to dismantle it. Now that the public is aware, I wouldn't be surprised if new golf courses start getting the same treatment.
50.
by daveguy 1mo ago
Opposition to AI data centers is completely rational, which is why both liberal and conservative folks are against it. These data centers are being forced on communities with soon-to-be-former leaders agreeing to nondisclosure agreements. T
51.
by daveguy 1mo ago
> its self-improvement loop ends where everyone’s does: “Humans review those suggestions as PRs on the factory definition and merge improvements." That's because it's still painfully clear that genAI has no taste. It'
52.
by daveguy 1mo ago
The problem with the down detector main reporting page is that all of the graphs are scaled to the same size. The OpenAI spike was nearly 40,000 and the Google spike was just over 100 (just over 400 for Gemini). They look the same in the re
53.
by daveguy 1mo ago
Wait. It's going to happen and it's SOTA, but you can't read the article because genAI produces such garbage? Seems like a disconnect there.
54.
by daveguy 1mo ago
By it do you mean us depending more and more on code that no one has read and understood?
55.
by daveguy 1mo ago
No one ever accused this administration of being competent.
56.
by daveguy 1mo ago
Yeah, I think AI is a great tool to get a handle on a codebase so you understand it. It's just a much more difficult task when that code is AI generated. I don't think agentic coding is worth all the overhead and complexity. Hones
57.
by daveguy 1mo ago
This is exactly how AI should be used when coding. "Agentic" AI produces garbage code. The entire codebase should never be accessible to the context (even if parts at a time, if the agent chooses the parts). And the entire codebas
58.
by daveguy 1mo ago
I've inherited plenty of code bases. AI generated code is a special kind of foobar. With a code base inherited from a human you can at least be assured at least one human has understood each portion for a brief period. With agentic slo
59.
by daveguy 1mo ago
I would rather see some awkward phrasing than bland LLM slop! Also, many plotting libraries include an xkcd style these days: https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.... And libraries for
60.
by daveguy 1mo ago
Well said. No one should trust an llm without verification/validation. They're just not good enough to do that and they have no sense of responsibility or ability to take it themselves. LLMs are empty.

More ›