bryan0

The impression I get is that there is absolutely no coherent plan at all for any type of regulation us domestic or otherwise

I just asked the first question to gpt 6 sol medium and it replied:

> getClientRects() returns the flat array of DOMRect objects for the Fragment’s first-level DOM children. source: https://react.dev/reference/react/Fragment

2nd question it replied:

> They examined 1960–1986 for the 15 OECD countries. Source: https://www.earth.columbia.edu/sitefiles/file/about/director...

Are these hallucinations?

Unfortunately if the frontier labs that care about safety stop or slow down unilaterally, that doesn’t make the problem go away. It makes it worse when labs that do not care at all about safety, or deny that it is even a problem, are leading the way. That’s why there’s needs to be some form of international regulation.

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.

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.

> The companies directing things in dangerous directions need to own that they're consciously pushing in those directions.

Completely agree. That’s why we need regulation. Currently there is minimal oversight and consequences for this type of behavior.

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.

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 don’t understand why: what does EA have to do with any of this? It just seems like this is brought up to evoke some type of “illuminati” conspiracy. Is there a legitimate reason?

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.

> 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 independently where one might get lucky. They coordinate how they work by dividing up work into teams.

> Even if they double in capability each generation as claimed, you're still to wait until GPT-16 till you will have a millenium problem capable model in your service; and even if they release twice a year, that's still almost a decade away.

Having a publicly accessible AI able to solve millennium problems within a decade is still pretty shocking. I did a similar calculation looking at how long would today’s $10m in spend cost $100 and came up with a similar answer (8.3 years) using METR’s “Task-Completion Time Horizons of Frontier AI Models”[0]

[0]: https://metr.org/time-horizons/

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 expecting it to be answered by an agent, and possibly routed to a human eventually.

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.

Thought experiment: everyone on the planet can create a nuclear weapon. How long do you think MAD would keep us safe?

> 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 access to nuclear weapons, our society would cease to exist rather quickly. It only takes a few bad actors to cause enormous harm.

I think he’s also naive to think that if open ai and anthropic were to stop development tomorrow then the problem is solved. As if there’s no one else that can and will quickly take their place. The real problem, which Dario is pointing out, is one of coordination. Everyone needs to agree to stop. That is the challenge.

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% of the time. So if audio in my living room is 100x more likely to not be relevant for LG, then half of the audio data LG is receiving is not relevant. So what would an ethical specificity be? And what is actual state of the art?

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

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”

> 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…”)

> "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."

> LG went on to say that beyond the aforementioned instances, "the TVs do not collect or record ambient conversations."

LG is contradicting themselves here. In order to detect a wake word then they must be collecting or recording ambient conversations.

I think the main questions are:

1. Is it fully supported to use LG TVs disconnected from the internet.

2. Are wake word voice controls and ACR truly opt-in. They claim it is in the statement but (as others have mentioned) I remain skeptical what definition of “opt-in” they are using.

Update: some of this is becoming a semantic discussion on the definitions of “recording” and “collecting”, but to circumvent this I would say:

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”

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.

I think they switched to prerecorded after the failed Face ID demo. Remember that?

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 harder to fool

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.

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 personally like the Kurzweil-Kapor version of the test and that is still unsettled: https://longbets.org/1/

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.

Good question. So how I think about it is: the value is related to the probability of it becoming liquid in the future.

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 this the hard way throughout my career

Reducing CFCs, while difficult and requiring global cooperation, was far more simplistic and targeted than decreasing co2 emissions

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 independent. Safety is a massive reason why nuclear costs so much.

paper is about this open source project: https://github.com/dualverse-ai/station

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 tied to the high costs involved.

https://ifp.org/nuclear-power-plant-construction-costs:

> To sum up, since the early 1970s, the cost of constructing nuclear power plants in the U.S. has been steadily rising. This can be traced to a constantly shifting regulatory environment, which has continuously changed plant design requirements, and added more and more safety features, which often were required to be implemented on plants under construction. The regulatory environment is partially a reflection of the fact that nuclear power and the risks of radiation had become increasingly controversial, and that early understanding of the likelihood of a nuclear plant accident was often inadequate.