Earlier quoted context omitted.
To be fair: a sizable proportion of humans are like that.
To be fair: that's just your sample. I don't see that.
Same. Most people afaict just live with little to no ambitions.
Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
691–700 of 754 posts
Re: Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
#692Earlier quoted context omitted.
Isn't that a massive quality improvement though? How many applications for LLMs are not feasible right now because of the ability for models to be swayed of course by a gentle breeze? If AI is a ship with a sail, data is the wind and alignment is the equivalent of a rudder.
The ability for models to be (easily) swayed is a different problem. I don’t see how AI safety would help with that.
> models to be (easily) swayed is a different problem
No, this is the alignment problem at a high level. You want a model to do X but sometimes it does Y.
Mechanistic interpretability, one area of study in AI alignment, is concerned with being able to reason about how a network "makes decisions" that lead it to an output.
If you wanted an LLM that doesn't succumb to certain prompt injections, it could be very helpful to be able to identity key points in the network that took the AI out of bounds.
Edit: I should add, I'm not referring to AI safety, I'm referring to AI alignment.
Re: Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
#693Earlier quoted context omitted.
Until I can trust that when I send an AI agent off to do something that I will be successful without me babysitting and watching over it constantly AI won't truly be transformative (since the human bottleneck will remain). This is one of the core promises of alignment. Without it how can there be trust? While there are probably short term slow downs with an alignment focus, ultimately it is necessary to avoid throwin…
I wouldn't mind a focus on reliably following tasks with greater intelligence; what I think is negative utility is focusing more compute and research resources on hypothetical superintelligence alignment — the entire focus of Ilya's "Superalignment" project — when GPT-4 is still way, way sub-human-intelligence. For example, I don't think the GPT Store was in any way a dangerous idea, which seems to have been Ilya's c…
I wouldn't call GPT-4 sub-human intelligence. While it's intelligence is less robust aggregate human intelligence, I don't think there is any one person alive who can compete with the breadth of GPT-4 knowledge.
I also think that the potential of what currently is possible with existing models has not been fully realized. Good prompting strategies and reflection may already be able to produce a system that is effectively AGI. Might already exist in several labs.
Re: Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
#694Earlier quoted context omitted.
These are FalungGong. I will not trust Falun Gong's news on China. They are known to create conspiracy stories.
I mean I would too if my ethnicity was so repressed, along with all the other non han Chinese.
Falun Gong is a religion, not an ethnicity, and they are of the cultish variety.
It's like believing the scientology center. Not trustworthy, they have an angle.
Re: Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
#695Jeremy Howard (of fast.ai): https://x.com/jeremyphoward/status/1725712220955586899 He is not exactly an insider, but seems broadly aligned/sympathetic/well-connected with the Ilya/researchers faction, his tweet/perspective was a useful proxy into what that split may have felt like internally.
> He is not exactly an insider, but seems broadly aligned/sympathetic/well-connected with the Ilya/researchers faction, his tweet/perspective was a useful proxy into what that split may have felt like internally.
Great analysis, thank you.
Re: Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
#696Earlier quoted context omitted.
As someone who has orchestrated two coups in different organizations, where the leadership did not align with the organization's interests and missions, I can assure you that the final stage of such a coup is not something that can be executed after just an hour of preparation or thought. It requires months of planning. The trigger is only pulled when there is sufficient evidence or justification for such action. Bui…
How are you still hireable? If I knew you orchestrated two coups at previous companies and I was responsible for hiring, you would be radioactive to me. Especially knowing that all that effort went into putting together a successful coup over other work. Coups, in general, are the domain of the petty. One need only look at Ilya and D'Angelo to see this in action. D'Angelo neutered Quora by pushing out its co-founder,…
I feel like in the parent comment coup is sort of shorthand for the painful but necessary work of building consensus that it is time for new leadership. Necessary is in the eye of the beholder. These certainly can be petty when they are bald-faced power grabs, but they equally can be noble if the leader is a despot or a criminal. I would also not call Sam Altman's ouster a coup even if the board were manipulated into ousting him, he was removed by exactly the people who are allowed to remove him. Coups are necessarily extrajudicial.
Re: Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
#697Earlier quoted context omitted.
If you want to get really basic: there's no OpenAI at all without Sam Altman, which means there's no ChatGPT either. There are much larger armies of highly qualified scientists and engineers at Google, Microsoft, Facebook, and other companies and none of them created ChatGPT. They wrote papers and did experiments but created nothing even remotely as useful. And they still haven't been able to even fully clone it with…
To pin OpenAI's success completely on Sam is disingenuous at best, outright dishonest at worst. Incentives don't build ML pipelines and infrastructure, developers and scientists do. This visionary bullshit is exactly that, bullshit.
A leader can't do anything on their own, they need people to lead. And those people deserve recognition and rewards. But in most cases there's no one more important than the leader. And thus, no one that deserves more credit than the leader.
I'm absolutely not comparing Sam Altman to any of these leaders, but just to illustrate how much vision and leadership does matter. Consider how stupid these statements sound:
"Jesus didn't build any churches, those were all built by brick layers and carpenters!"
"Pharaohs didn't build a single pyramid, those were all built by artists and workers!"
"Abraham Lincoln didn't free any slaves, he didn't break the chains of a single slave, that was all done by blacksmiths!"
"Martin Luther King Jr. didn't radically improve civil rights, he never passed a single law, that was all done by lawmakers!"
"Sam Altman didn't build ChatGPT, he didn't create a single ML pipeline, it was all done by engineers!"
It's a hard fact of life that some specific individuals play more important roles in successful projects than others.
Re: Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
#698Earlier quoted context omitted.
> The board's job is not to do right by 'Sam & Greg', but to do right by OpenAI. The board's job is specifically to do right by the charitable mission of the nonprofit of which they are the board. Investors in the downstream for-profit entity (OpenAI Global LLC) are warned explicitly that such investments should be treated as if they were donations and that returning profits to them is not the objective of the firm,…
> charitable mission of the nonprofit of which they are the board This exactly. Folks have completely forgotten that Altman and Co have largely bastardized the vision of OpenAI for sport and profit. It's very possible that this is part of a larger attempt to return to the stated mission of the organization. An outcome that is undoubtedly better for humanity.
Have they though..
What is the evidence of that, and what is your evidence that this "return to mission" will "undoubtedly better for humanity"
After all, as we see by looking to history, the road to hell is paved with good intentions, lots and lots of altruistic do gooders have created all manner of evil in their pursuit of a "better humanity".
I am not sure I agree with Sam Altman's vision of a "better tomorrow" any more than I would agree with the OpenAI boards vision of that same tomorrow. In fact I have great distrust of people that want to shape humanity into their vision of what is "best" that tends to lead to oppression and suffering
Re: Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
#699Earlier quoted context omitted.
Won't a truly open model conflict with the AI executive order?
What do you mean by AI executive order?
Re: Details emerge of surprise board coup that ousted CEO Sam Altman at OpenAI
#700Earlier quoted context omitted.
No, literal attempts. One attempt involved a battleship “accidentally” firing onto another battleship where both Hu Jintao and Xi Jinping were visiting. https://jamestown.org/program/president-xi-suspects-politica... Biased source, but she’s able to get a lot of unreported news from the mainland. https://www.jenniferzengblog.com/home/2021/9/20/deleted-repo... I will try to find more sources but Google is just shit th…
These are FalungGong. I will not trust Falun Gong's news on China. They are known to create conspiracy stories.
1. The sources aren’t limited to falun gong
2. It makes sense given Xi’s current paranoia and constant purges