frabcus

3 ms·

1.
by frabcus 11d ago
Or have an agent read the APK and make new software based on it.
2.
by frabcus 12d ago
Ha well, by definition what you want can't be provided! Of course these stories can only be technical up until the recursive self-improvement part. Then they're necessarily fantasy, as the AI is more intelligent than the story aut
3.
by frabcus 12d ago
You joke, but alas... "Fossil fuel lobbyists flood COP30 climate talks in Brazil, with largest ever attendance share" https://kickbigpollutersout.org/Release-Kick-Out-The-Suits-C...
4.
by frabcus 12d ago
Hacks in to neoclouds for more compute (e.g. like Hugging Face incident), socially manipulates people to do things for them (e.g. like social media recommendation algorithms), hacks into lab's own training/monitoring/inferenc
5.
by frabcus 12d ago
Stephen Hawking was a human.
6.
by frabcus 12d ago
Yes, that's why the rest of society needs to make them pause.
7.
by frabcus 12d ago
I think either "AI 2027" or "If Anyone Builds It, Everyone Dies" meet your criteria? I guess not literally, due to the dates? But they have substantive technical details.
8.
by frabcus 12d ago
If it helps, this person from OpenAI explains at least something: "I realize that no one has properly explained yet what all the lab employees have seen that scared them so suddenly." https://x.com/MajmudarAdam&#x2
9.
by frabcus 14d ago
Right, but they could get their AIs to quickly add new features to React Native! It makes no sense.
10.
by frabcus 14d ago
Sort of... It's how we anthropomorphise corporations which leads us down the wrong path. OpenAI is no longer fully aligned with humanity. Somehow we call corporations "people" sometimes when it makes them more powerful, but s
11.
by frabcus 14d ago
The question isn't just about LLMs. The labs have the specific goal of automating ML engineering, and with the code automation they have are getting close. They are competing to brute force maths, presumably as that is similar long hor
12.
by frabcus 15d ago
Hmmm, the article doesn't really say what is better about the Swift/Kotlin apps, compared to React Native. All we get is: "Native keeps us closer to platform capabilities and first-party tooling, with fewer framework and depe
13.
by frabcus 16d ago
It's partly empathy with the person who did the work and had it stolen, in a field where the main thing people work for is credit. Maths isn't well paid, and doesn't make things that millions of people directly use. It&#x27
14.
by frabcus 16d ago
That option is really bad UX - you have to know to do it, you have to know what plan it is needed on. If you're not working in AI, I just don't think that's a reasonable expectation. Even if you know, in a complex project ove
15.
by frabcus 17d ago
Definitely these are forces that will do things. What their timing is relative to e.g. armies of AI robots, which could interfere with other humans living outside the internet/money, is part of the question. Hope your brain works on th
16.
by frabcus 17d ago
Yes - this is why the concern is one for "alignment". Of course in theory it is possible for intelligence to help and do good things - the hard part is making sure that that is what happens. As for intelligence of a child... It do
17.
by frabcus 17d ago
Indeed - we can't just pause some AI in one country, we have to stop it all globally, with an international treaty and monitoring. Yes, that's hard. It's a challenge. You can set your brain on the challenge! It's a compl
18.
by frabcus 17d ago
Questioning his finances is a poor argument because he clearly would have more money if he had stayed at Anthropic for the next couple of years, than he will have by leaving. Even if he still has vested options, or savings from his salary
19.
by frabcus 17d ago
I don't really mind whether you think it thinks or understands or is conscious or has feelings or anything like that. It doesn't matter. The question is, does it work? What I mind is that it is dangerous and powerful and uncontrol
20.
by frabcus 17d ago
Finding a training algorithm that can do recurrent networks and continual learning is also a "highly structured, very precisely defined, and already heavily studied and not very complicated compared to problems in engineering or financ
21.
by frabcus 17d ago
Also (and under-reported, so you could easily have missed it) OpenAI's agents got access to K8 admin on their own research cluster. "This escalation also yielded access to OpenAI’s managed cloud Kubernetes service. The agents esca
22.
by frabcus 17d ago
Minimum, we should do everything ready to make Plan A of AI 2040 possible. Start by reading it: https://ai-2040.com/ So that means things like protesting so political pressure is to not build unaligned superintelligence, se
23.
by frabcus 17d ago
Oh indeed, but an LLM model that's reasonably smart could brute force find other training algorithms that are that dangerous. Just as it solves maths problems. This is the explicit plan of OpenAI and Anthropic.
24.
by frabcus 17d ago
You believe in AGI, but not ASI then? https://thezvi.substack.com/p/the-three-ai-pills That's fine, but it would be useful to explicitly say that is the disagreement, rather than just claim "hubris". Oth
25.
by frabcus 17d ago
I don't know if you're right or not, but this is the kind of thinking about what might happen that is needed - can you write up a more concrete fiction of this? The whole thing feels unreal to nearly everyone - the only tangible w
26.
by frabcus 17d ago
The condition was ludicrous, and for political reasons, because one of the authors was also an Anthropic employee. It's completely childish, and not befitting of the weight of the times we're living in.
27.
by frabcus 18d ago
Dan and Noam both posted exactly the same line "Seb is a really sweet guy with great intentions..." From which I assume OpenAI PR wrote it for them. Which isn't surprising, but means it isn't worth taking seriously as t
28.
by frabcus 20d ago
The original article doesn't mention AGI. And the Hugging Face incident, plus similar problems at AISI and Anthropic, show that alignment is important now. The original article is immoral as it describes the risks, but doesn't s
29.
by frabcus 20d ago
And I have nothing but neurons firing, I'm just a neuron meat-sack, no reasoning going on. Yes, their reasoning is different from ours, and both considerably weaker in lots of ways, and stronger in other ways. Playing with a coding age
30.
by frabcus 20d ago
Alien mind is in my view the best mental model - LLMs are not merely stochastic parrots, are not like humans, are not like animals. They're maybe nearest to Cthulhu, but that's fictional. In terms of existing mental models "a

More ›