daveguy

12 ms·

1.
by daveguy 3d ago
This makes more sense. Automatic loss for the offending parent.
2.
by daveguy 3d ago
Result is a permanent cancellation? Does that mean the match is nullified or the winner at the time wins? Either way, one team could draw a strategic black card. Seems like cancellation of the match is not the correct outcome. What about p
3.
by daveguy 3d ago
But.... Money.
4.
by daveguy 3d ago
Wait... Which of the 2 did Altman not do where Amodei did? I mean, maybe not quite 8 attempts at regulatory capture. But come on. And the only reason OpenAI didn't get blacklisted is they rolled over for the worst applications.
5.
by daveguy 3d ago
One question: did you jump straight from programming to depending on the AI without looking at the code? If so, I've got some bad news for you.
6.
by daveguy 3d ago
The illusions of thinking produced by a "thinking trace" is just as ignorant of reality as the first and last tokens. All they know of reality is the tokens in their context.
7.
by daveguy 3d ago
The problem is, LLMs have such little understanding of the world around them. "Find exploits in specific software on this device" may as well be "find exploits".
8.
by daveguy 4d ago
I would love for good governance principles to seep into corporations rather than the other way around. Seems about time the direction reversed.
9.
by daveguy 4d ago
Or you might find out how little people care about AI. Which would not bode well for bearing load.
10.
by daveguy 4d ago
Yeah, vibe coding produces shit software.
11.
by daveguy 5d ago
The point isn't what happens after something is solved. Where is the motivation for humans to study/find a potential solution if your kids are going to starve while you do it? These things can brute force problems that have a cle
12.
by daveguy 5d ago
It also takes human labor to run an effective nigerian prince scam. That doesn't make it okay.
13.
by daveguy 5d ago
But also, fuck trump.
14.
by daveguy 5d 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.
15.
by daveguy 6d ago
The coming shitstorm of crappy software is going to suck.
16.
by daveguy 6d ago
Nah, that just makes it a shitty tool.
17.
by daveguy 7d ago
Slick bullshit. That's AI.
18.
by daveguy 7d ago
I don't think they meant the nurse would be responsible. The LLC would and should be fucked in that case.
19.
by daveguy 8d 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".
20.
by daveguy 8d 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
21.
by daveguy 8d 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
22.
by daveguy 10d 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.
23.
by daveguy 10d 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
24.
by daveguy 10d 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.
25.
by daveguy 11d ago
But understanding that requires reading the code or writing that it spits out rather than having yet another LLM do it for you.
26.
by daveguy 11d 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
27.
by daveguy 11d 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.
28.
by daveguy 11d 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
29.
by daveguy 12d 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
30.
by daveguy 12d 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

More ›