bryan0

3 ms·

1.
by bryan0 6d ago
It is sad that this is the best level of reporting and oversight of these types of events available to us, but it is currently all we have. We have to do better, but to dismiss the report because of the style and tone would be foolish.
2.
by bryan0 6d ago
Just because what happened is a “direct consequence of risky human choices” that does not make it any less dangerous. Agents will eventually cause serious harm to online infra whether it’s intentionally human-directed or accidental. > Th
3.
by bryan0 6d ago
It really doesn’t though. It criticizes how the news media reported on the incident but this opinion piece is no better. Just read the original source itself.
4.
by bryan0 6d ago
It’s not a majority opinion because you have to do some serious mental gymnastics to turn this demonstration of dangerous AI behavior into a PR publicity stunt. Serious question though because I’ve seen this brought up several times and I d
5.
by bryan0 6d ago
I would just recommend reading what actually happened: https://metr.org/blog/2026-08-26-openai-hugging-face-inciden... I don’t think downplaying what occurred is really beneficial to anyone.
6.
by bryan0 8d ago
> The models are really impressive, but I don't think it's cope to remark that this was in essence a 1:10000 chance outcome. I don’t think that’s an accurate way to think about what’s going on. These agents are not acting indep
7.
by bryan0 11d ago
What does "contacting another human" actually mean when all of these messages are being answered by agents first? If I'm texting my friends and family I expect that to be answered by a human, but beyond that I am partially ex
8.
by bryan0 11d ago
meanwhile agents have been able to pass these captchas for a while now, so it's unclear what the point of them now is. It's like they want to keep out the dumb bots, but any agent with reasonable intelligence can come in.
9.
by bryan0 13d ago
Thought experiment: everyone on the planet can create a nuclear weapon. How long do you think MAD would keep us safe?
10.
by bryan0 13d ago
> A powerful technology that is out there for everyone to use comes with built-in pacing. In a way it’s the truest form of MAD or proliferation. I think this misunderstanding of MAD undermines his entire point. If everyone had equal acce
11.
Biggest AI Rivals Agree They Need to Slow It Down(wsj.com)
2 points by bryan0 13d ago|0 comments
12.
by bryan0 13d ago
Point taken. Let’s focus on the “ethical implementations do exist” part though. Let’s say a best possible implementation has 99% specificity. Then if it detects audio that has nothing to do with “LG” it mistakenly treats it as LG relevant 1
13.
by bryan0 14d ago
Dario has been pretty clear and consistent that: 1. AI model safety is the most important thing. 2. open weights decreases model safety So it seems unlikely that this suggestion would be well-received
14.
by bryan0 14d ago
Fair points, but if a device can detect words in my ambient conversations and depending on what it detects (accurate or not) it can send that audio to the cloud, I would describe that device as “collecting or recording ambient conversations
15.
by bryan0 14d ago
> Assume it works perfectly accurately. Well this is a silly assumption. Wake word false positives happen all the time. (“No Siri, I wasn’t talking to you…”)
16.
by bryan0 14d ago
> "LG TVs process voice data only when the voice button on the remote control is pressed and held, or when a wake word such as 'Hi LG' is recognized after the user has activated the Far-Field voice recognition feature.&quo
17.
by bryan0 15d ago
One proposal that Tao hints at is to not rush to announce solutions. Instead maybe the AI companies should work privately with the subject matter experts on how to communicate the discoveries.
18.
by bryan0 16d ago
I think they switched to prerecorded after the failed Face ID demo. Remember that?
19.
Finite-time blowup with smooth forcing for incompressible porous media(mastodon.social)
2 points by bryan0 18d ago|0 comments
20.
by bryan0 22d ago
Try it. It’s really not that easy. The other thing is that the judges would be probing it with jailbreaks like “ignore previous instruction” attacks. You could actually probably have llm judges at this point which might be ironically even h
21.
by bryan0 22d ago
Definitely would not be easy. First of all the mainstream llms are trained to be honest, and this requires lying convincingly. Second, this involves 8 hours of interviews with expert judges, one "claudism" could give it away.
22.
by bryan0 23d ago
that would be a reasonable definition of AGI if everyone agree upon the specifics of the test, but that has never happened. Turing test is very much out of style, but I think that's because no one could even agree what the test was. I
23.
by bryan0 23d ago
There is no world in which it makes sense to take this risk. There are plenty of great alternatives where you don't have to worry about playing russian roulette with your entire online identity.
24.
by bryan0 26d ago
Good question. So how I think about it is: the value is related to the probability of it becoming liquid in the future.
25.
by bryan0 27d ago
Oh yeah that’s easy! Why didn’t he think of that earlier? /s For those who don’t know, just because you have a valuable asset, e.g. stock in a private company, that does not necessarily mean you can sell it for cash. I’ve experienced t
26.
by bryan0 27d ago
Reducing CFCs, while difficult and requiring global cooperation, was far more simplistic and targeted than decreasing co2 emissions
27.
Employers Are Making Job Candidates Jump Through Hoops to Prove They're Real(wsj.com)
2 points by bryan0 27d ago|1 comments
28.
by bryan0 28d ago
Oh ok, yeah I think we agree. the original quote I was responding to was "the (false) idea that what was holding back nuclear was perception of safety, rather than cost." I was simply trying to say that safety and cost are not ind
29.
by bryan0 28d ago
paper is about this open source project: https://github.com/dualverse-ai/station
30.
by bryan0 28d ago
I don't think I get your point. A project will not be approved and funded if the local population does not believe it is safe, so safety must be demonstrated through a variety of means, including some you mentioned. This is directly ti

More ›