Phoenix Beaudry

@Metathea11 @KojimaErgoSum @tszzl Everything ought to be explained as simple as possible - but no simpler.

@KojimaErgoSum @tszzl That is a vast simplification though It’s all relative

i think reveling in complexity is very anti scientific, anti technological. i hate it when people draw some giant metabolic or ecological diagram to demonstrate how well they’ve understood it. all that tells me is that we’re nowhere near understanding it

bittensor.com/documentation/… minimalistic style meets dense knowledge

@goafterwork @Techweek_ first time in SF hope there is more spots available for the {after} party :)

Who wants to party with us during #sftechweek 🥳 partiful.com/e/kkeJq1bfzQiK… @Techweek_

Alignment should probably begin and end with "don't kill everyone", let it learn the rest Most would not appreciate an immortal super-entity aligned to 1720s human values Would 2200 appreciate one aligned to 2020s human values?

@RandolphCarterZ @made_by_chris However the kind of super-optimizer I am thinking of is ignorant of it's own reward function so maybe capping these systems is sufficient.

@RandolphCarterZ @made_by_chris You don't think it's possible that super-intelligence and super-optimizers are different evolutionary paths and we might accidentally find ourselves on the wrong one?

Anti-doom alignment: Put a cap on reward function That's it Turn "Universe into paperclips" into "A lot of paperclips" with this one simple trick

@RandolphCarterZ Is there a reason that is a ridiculous thing to think might happen?

Why do people keep talking about AI "Escaping into the internet"? What is the actual fear here - that they could buy or steal some cloud credits and run multiple instances of themselves?

If you grant social network apps access to “All Photos” ⚠️ … they can listen to the spoken Audio 💬 in your Live Photos! 🔊 … My friend was just chatting to me about how this is one way ad targeting software can “listen” 🔮 without showing the Mic indicator on your phone. 😱

Do you know what is going on inside the pupa?

new research from OpenAI used gpt4 to label all 307,200 neurons in gpt2, labeling each with plain english descriptions of the role each neuron plays in the model. this opens up a new direction in explainability and alignment in AI, helping make models more explainable and potentially easier to align. openai.com/research/langu…

@OpenAI the debate rates

(I still think AI risk is a huge deal, by the way. I suspect some people think I'm a total AI risk skeptic, but I'm not. I just don't give human extinction from AI >50% chance.)

In this post @robbensinger neatly lays out the argument for a very high likelihood of AI doom. I probably disagree with premise 2b the most. It depends on what counts as "full value loading" but I've not been persuaded that it's "~impossibly difficult". lesswrong.com/posts/QzkTfj4H…

This Tweet was automatically sent just by asking Calo to send a tweet :) stay tuned for the beta.

here is an alternative path for society: ignore the culture war. ignore the attention war. make safe agi. make fusion. make people smarter and healthier. make 20 other things of that magnitude.   start radical growth, inclusivity, and optimism.   expand throughout the universe.

the creative power of the world is breathtaking to watch. each time our tools get a little better and a little bit more accessible, our creations improve more than we imagine possible.

@leecronin Where do LLMs fall on the Assembly theory complexity scale? Where do systems of multiple recurrent LLMs that interact with each other?

LLMs are not sentient in any way shape or form. The data was produced by sentient beings so any feelings of that is pass through or anthropomorphic hallucinations.

This is Art