What was her angle from beginning?
Mira Murati leaves OpenAI
291–300 of 644 posts
Re: Mira Murati leaves OpenAI
#292Hopefully she didn't generate her farewell message using AI.
Re: Mira Murati leaves OpenAI
#293Earlier quoted context omitted.
[flagged]
Is that your test suite?
Companies are held to the standard that their leadership communicates (which, by the way, is also a strong influencing factor in their valuation). People don't lob these complaints at Gemini, but the CEO of Google also isn't going on podcasts saying that he stares at an axe on the wall of his office all day musing about how the software he's building might end the world. So its a little understandable that OpenAI would be held to a slightly higher standard; its only commensurate with the valuation their leadership (singular, person) dictates.
Re: Mira Murati leaves OpenAI
#294It is hard for me to square "This company is a few short years away from building world-changing AGI" and "I'm stepping away to do my own thing". Maybe I'm just bad at putting myself in someone else's shoes, but I feel like if I had spent years working towards a vision of AGI, and thought that success was finally just around the corner, it'd be very difficult to walk away.
> It is hard for me to square "This company is a few short years away from building world-changing AGI" Altmans quote was that "it's possible that we will have superintelligence in a few thousand days", which sounds a lot more optimistic on the surface than it actually is. A few thousand days could be interpreted as 10 years or more, and by adding the "possibly" qualifier he didn't even really commit to that predicti…
Because as we all know: Full Self Driving is just six months away.
Re: Mira Murati leaves OpenAI
#295They can't spend more than 6 months without a drama...
It's the same drama, spread out over time.
Re: Mira Murati leaves OpenAI
#296> we fundamentally changed how AI systems learn and reason through complex problems
I'm not an AI researcher, have they done this? The commentary I've seen on o1 is basically that they incorporated techniques that were already being used.
I'd also be curious to learn: what fundamental contributions to research has OpenAI made?
The ChatGPT that was released in 2022 was based on Google's research, and IMO the internal Google chatbot from 2021 was better than the first ChatGPT.
I know they employ a lot of AI scientists who have previously published milestone work, and I've read at least one OpenAI paper. But I'm genuinely unaware of what fundamental breakthroughs they've made as a company.
I'm willing to believe they've done important work, and I'm seriously asking for pointers to some of it. What I know of them is mainly that they've been first to market with existing tech, possibly training on more data.
Re: Mira Murati leaves OpenAI
#297Earlier quoted context omitted.
To be fair, that question is one of the suggested questions that OpenAI shows themselves in the UI, for the o1-preview model. (Together with 'Is a hot dog a sandwich?', which I confess I will have to ask it now.)
If you have a sandwich and cut it in half, do you have one or two sandwiches?
Depends on what kind of sandwich it was before, and along which axis you cut it, and where you fall on the sandwich alignment chart.
Re: Mira Murati leaves OpenAI
#298Sutskever [1], Karpathy [2], Schulman [3], and Murati today! Who's next? Altman ?! _________________ 1. https://news.ycombinator.com/item?id=40361128 2. https://news.ycombinator.com/item?id=39365935 3. https://news.ycombinator.com/item?id=41168904
I don't think Altman will push himself out a window.
Re: Mira Murati leaves OpenAI
#299It is hard for me to square "This company is a few short years away from building world-changing AGI" and "I'm stepping away to do my own thing". Maybe I'm just bad at putting myself in someone else's shoes, but I feel like if I had spent years working towards a vision of AGI, and thought that success was finally just around the corner, it'd be very difficult to walk away.
Maybe it has to do with Sam getting rid of the nonprofit control and having equity?
Re: Mira Murati leaves OpenAI
#300Earlier quoted context omitted.
Her rise didn't make sense to me. Product manager at tesla to CTO at openAI with no technical background and a deleted profile ? This is a very strange company to say the least.
A significant portion of the old guard at OpenAI was part of the Effective Altruism, AI Alignment, and Open Philanthropy movement. Most hiring in the foundational AI/model space is very nepotistic and biased towards people in that clique. Also, Elon Musk used to be the primary patron for OpenAI before losing interest during the AI Winter in the late 2010s.
Which has zero explanatory power w.r.t. Murati, since she's not part of that crowd at all. But her previously working at an Elon company seems like a plausible route, if she did in fact join before he left OpenAI (since he left in Feb 2018).