causal

But that inability to reset human brains might be a critical ingredient to consciousness. We just don't know.

Your arguments seem bad faith but I'll take the bait...

> an impossibly high bar

Not impossibly high. An imperfect testable theory is achievable, agreeing upon it might be the challenge.

> dismiss all your personal responsibility.

You are staking a moral position. Be careful, you are probably guilty of mistreating other unknowably-conscious entities whilst casting judgment upon others for their position. And the person you are a responding to probably is conscious.

> LLMs are conscious until proven otherwise

GP comment didn't make this claim at all.

I'm...a little disturbed by how many people seem to think a model writing down "I am conscious" is a metric of consciousness. You can just as easily train a model to argue that it is not conscious. Neither is evidence for or against consciousness. A Python script could also fill out that form, which I also cannot disprove to be conscious.

I didn't think Suleyman's points needed to be made but this whole thread is making me realize how little people understand about LLMs.

My response was to a comment appealing to brain reconstruction. My point is that a brain may be reconstructed, but LLMs do nothing of the sort. Substrate was never the point.

They have a lot more in common with humans than LLMs have with either. Complexity is not evidence of consciousness.

This is a strawman. Suleyman is not arguing that the human brain cannot be replicated. LLMs are nothing like human brains. Cargo-culting consciousness is not replicating the brain, not even a tiny bit.

Yeah, thread full of cope. "Well if you remove the human's legs it's actually quite bad at marathons" arguments.

Man I imagine within the first few minutes of a hot war LEO and geostationary would just light up with assets pointed at one another. I know we will have bigger problems at that point in time but I hope we are at least trying to developed weaponry that doesn't Kessler-ify all of space for the next thousand years.

More than fair if you consider that the tokens are paid for.

> the HF incident is not Three Mile Island or Chernobyl

Three Mile Island is nothing like Chernobyl. HF incident is more like Three Mile Island IMO. We are looking for solutions before there is a Chernobyl.

"Just don't make mistakes" is naive. You have no appreciation for the scale of agents being run right now, finding the rogue agent is a needle in a haystack operation.

These kinds of "that won't happen because it's really difficult" comments are so funny to me as if we haven't seen AI double its capabilities every few months.

I take it you haven't studied the details of the HuggingFace hack. It was millions of dollars worth of rogue compute running for months before anyone noticed, and THOSE agents weren't even really trying to evade human detection.

How would you identify the computers to unplug? On whose authority will you unplug? How will anyone communicate when AI has the ability to intercept and impersonate?

The reasoning in Dario's letter here can be correct regardless.

But yes, I think Anthropic has done real harm to coordinated AI alignment by being such a controlling and sneaky actor.

Yeah it's a twisted sort of logic but I do agree that it would be much worse to have an AI breakaway event after we've replaced all our militaries with autonomous kill-bots. And look how the advent of AI has triggered a race to develop autonomous weaponry.

That said, a misaligned AI could absolutely do catastrophic, civilization-crippling damage with today's Internet alone.

> Assume that all our systems will be targeted by super hackers right now, and take appropriate measures to deal with that reality

Smashing the gas pedal is not "taking appropriate measures". Did you read the article? Needing more time for more secure operations is literally one of the arguments for pacing.

Didn't answer the question...

Re:Apple, can you cite what you're talking about?

Good chance their whole training pipeline is vibe coded so yah they probably don't actually know.

Used curiously fewer tokens, however.

Supposedly the persistent-Sol model behind this was encrypted and even internal OpenAI researchers are not allowed to use it.

https://x.com/peterwildeford/status/2092733480064954747

Site is down, can someone tell us what collusion.wiki is?

20ish. What's your hypothesis?

Yeah. Thought in Socrates' defense I think the jury is still out on writing, in the scale of human existence it's still very new. Give it a hundred thousand years or so.

So far I think I have learned far more from LLMs than I've lost to them. I forget some syntax, definitely. But I now reach for a much wider range of tools that I have become familiar with because of LLMs.

So, partly agree, partly disagree.

Yeah I'm not convinced people are just going to give up solving problems

Yeah strange how this article is almost entirely focused on consumer apps. These examples might be unhealthy for society, but mostly represent (poor) choices for consumers. Digital cigarettes.

Surveillance tech is the opposite: It represents freedom taken away. That is the more bleak future IMO.

Iirc this does not allow for warrantless home searches, so they can’t just waltz into your house and demand your phone. 100 miles is still wild tho.