ASK Sathvik

MLguy who loves Math, Phy, Bio.
I want to live forever to understand the universe

[ASK Sathvik @AskSathvik]

@PradyuPrasad Why can’t we just ask the models regarding this?

[ASK Sathvik @AskSathvik]

@usr_bin_roygbiv who are leaving anthropic?

[ASK Sathvik @AskSathvik]

@gaxrav we don't have the hardware to see UV rays as well, but we can make equipment to do so.

[ASK Sathvik @AskSathvik]

@krisborer I can’t DM you because you aren’t following me. I pinged you on lbf slack.

[ASK Sathvik @AskSathvik]

@tszzl No way you are staying at OpenAI by talking as such, when are you leaving?

[ASK Sathvik @AskSathvik]

@PavanKatta1998 Why? What’s special about a finetuned model working well on a specific task?

[ASK Sathvik @AskSathvik]

@tszzl @nickcammarata @leothecurious the suffering caused consciously is the problem.

[ASK Sathvik @AskSathvik]

@tszzl @nickcammarata @leothecurious you are talking as if this was what the process was aiming for. It just happened in that way.

[ASK Sathvik @AskSathvik]

How will you know if someone is actually going to use the outputs of your model for distillation?

[Bill Gurley @bgurley]

If you are on the verge of AGI or ASI, why isn’t your model smart enough to recognize espionage distillation in real time? You say “cure cancer in a few years.” Isn’t sniffing illicit distillation quite a bit easier than curing cancer? Why write letters to DC? Just use AGI.

[ASK Sathvik @AskSathvik]

@ToKTeacher What should this model do that convinces one that it’s AGI. What’s truly novel?

[ASK Sathvik @AskSathvik]

Beyond the simple profit maximisation logic, Anthropic thinks that they are the only ones who can build aligned AGI, so they can justify anything they do with that logic.

[elie @eliebakouch]

mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community

also the fact that this is un purpose not visible to the user is crazy

[ASK Sathvik @AskSathvik]

@ToKTeacher understood, thanks for explaining this. I understand your argument more clearly now.

[ASK Sathvik @AskSathvik]

@ToKTeacher I am not saying that humans are fundamentally constrained that they cannot do whatever AGI can do.
But given our current biological constraints we will be much slower than AGI, even with the same set of tools.
Do you agree with the speed argument?

[ASK Sathvik @AskSathvik]

@ToKTeacher Sure, but a calculator doesn’t work independent of a user, that’s not true of AGI. You are assuming you can control it to do whatever you want, that assumption is doing a lot of work for your explanation.

[ASK Sathvik @AskSathvik]

Rather simplistic of him to assume that these thinking machines would have no will of their own

[blue @bluewmist]

I'm finally reading Dune. This quote, which is in the first few pages, hits hard:

"Once men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them."

[ASK Sathvik @AskSathvik]

@bluewmist Rather simplistic of him to assume machines have no will of their own

[ASK Sathvik @AskSathvik]

@RifeWithKaiju @fabianstelzer Thanks

[ASK Sathvik @AskSathvik]

@RifeWithKaiju @fabianstelzer I see you have written a paper, I will read it and reach out. How do I reach out to you @RifeWithKaiju DMs aren't open. mail?

[ASK Sathvik @AskSathvik]

@RifeWithKaiju @fabianstelzer At this point it’s just repeating stuff, so I’ll stop here.