Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

OpenAI departures: Why can’t former employees talk?

vox.com

Post reply on HN
Print view View original

131–140 of 1001 posts

Re: OpenAI departures: Why can’t former employees talk?

#131

Post by heavyset_go »

N0b8ez wrote: post #90
refulgentis wrote: post #63

Earlier quoted context omitted.

Adding a disclaimer for people unaware of context (I feel same as you): OpenAI made a large commitment to super-alignment in the not-so-distant past. I beleive mid-2023. Famously, it has always taken AI Safety™ very seriously. Regardless of anyone's feelings on the need for a dedicated team for it, you can chalk to one up as another instance of OpenAI cough leadership cough speaking out of both sides of it's mouth as…

>To really stress this: OpenAI's still-present cofounder shared yesterday on a podcast that they expect AGI in ~2 years and ASI (superpassing human intelligence) by end of the decade. Link? Is the ~2 year timeline a common estimate in the field?

We can't even get self-driving down in 2 years, we're nowhere near reaching general AI.

AI experts who aren't riding the hype train and getting high off of its fumes acknowledge that true AI is something we'll likely not see in our lifetimes.

Re: OpenAI departures: Why can’t former employees talk?

#132

Post by xyst »

ryandrake wrote: post #33

Non-disparagement clauses seem so petty and pathetic. Really? Your corporation is so fragile and thin-skinned that it can't even withstand someone saying mean words ? What's next? Forbidding ex-employees from sticking their tongue at you and saying "nyaa nyaa nyaa?"

The company is literally a house of cards at this point. There is probably so much vulture capitalist and angel investor money tied up in this company that even a disparaging rant could bring the whole company crashing down.

It’s yet another sign that the AI bubble will soon burst. The laughable release of “GPT-4o” was just a small red flag.

Got to keep the soldiers in check while the bean counters prep the books for an IPO and eventual early investor exit.

Almost smells like a SoftBank-esque failure in the near future.

Re: OpenAI departures: Why can’t former employees talk?

#133

Post by lopkeny12ko »

What a lot of people seem to be missing here is that RSUs are usually double-trigger for private companies. Vested shares are not yours. They are just an entitlement for you to be distributed common stock by the company. You don't own any real stock until those RSUs are released (typically from a liquidity event like an IPO).

Companies can cancel your vested equity for any reason. Read your employment contract carefully. For example, most RSU grants have a 7 year expiration. Even for shares that are vested, regardless of whether you leave the company or not, if 7 years have elapsed since they were granted, they are now worthless.

Re: OpenAI departures: Why can’t former employees talk?

#135

Post by DesiLurker »

apsec112 wrote: post #27
jp57 wrote: post #21

The only way I can see this being a valid contract is if the equity grant that they get to keep is a new grant offered the time of signing the exit contract. Any vested equity given as compensation for work could not then be offered again as consideration for signing a new agreement. Maybe the agreement is "we will accelerate vesting of your unvested equity if you sign this new agreement"? If that's the case then it…

It's not. The earlier tweets explain: the initial agreement says the employee must sign a "general release" or forfeit the equity, and then the general release they are asked to sign includes a lifetime no-criticism clause.

somebody explained to me early on that you cannot have a contract to have a contract. either initial agreement must state this condition clearly or they are signing another contract at employment termination which is bringing these new terms. IDK why would anyone sign that at termination unless they dangle additional equity. I dont think this BS they are trying to pull would be enforceable at least in California. though IANAL obviously.

all this said, in bigger picture I can understand not divulging trade secrets but not being allowed to discuss company culture towards AI safety essentially tells me that all the Sama talk about the 'for the good of humanity' is total BS. at the end of day its about market share and bottom line.

Re: OpenAI departures: Why can’t former employees talk?

#136

Post by riwsky »

jprete wrote: post #123
beastman82 wrote: post #101

Earlier quoted context omitted.

ITT: a bunch of laymen thinking their 2 second proposal will outlawyer the team of lawyers who drafted these.

Lawyers are 100% capable of knowingly crafting unenforceable agreements.

You don’t need to out-litigate the bear,

Re: OpenAI departures: Why can’t former employees talk?

#137

Post by timmg »

rvz wrote: post #18

So that explains the cult-like behaviour months ago when the company was under siege. Diamond multi-million dollar hand-cuffs which OpenAI has bound lifetime secret service-level NDAs which are another unusual company setting after their so-called "non-profit" founding and their contradictory name. Even an ex-employee saying 'ClosedAI' could see their PPUs evaporate in front of them to zero or they could never be all…

I don’t have any idea what goes on inside OAI. But I have this strange feeling that they were right to oust sama. They didn’t have the leverage to pull it off, though.

Re: OpenAI departures: Why can’t former employees talk?

#138

Post by DebtDeflation »

jp57 wrote: post #21

The only way I can see this being a valid contract is if the equity grant that they get to keep is a new grant offered the time of signing the exit contract. Any vested equity given as compensation for work could not then be offered again as consideration for signing a new agreement. Maybe the agreement is "we will accelerate vesting of your unvested equity if you sign this new agreement"? If that's the case then it…

My initial reaction was "Hold up - your RSUs vest, you sell the shares and pocket the cash, you quit OpenAI, a few years later you disparage them, and then when? They somehow try and claw back the equity? How? At what value? There's no way this can work." Then I remembered that OpenAI "equity" doesn't take the form of an RSU or option or anything else that can be converted into an actual share ever. What they call "equity" is a "Profit Participation Unit (PPU)" that once vested entitles you to a share of their profits. They don't share the equivalent of a Cap Table with employees, so there's no way to tell what sort of ownership interest a PPU represents. And of course, it's unlikely OpenAI will ever turn a profit (which if they did would be capped anyway). So this is all just play money anyway.

Re: OpenAI departures: Why can’t former employees talk?

#139

Post by ndriscoll »

ihumanable wrote: post #130
ndriscoll wrote: post #116

Earlier quoted context omitted.

That doesn't really contradict what the other poster said. They're calling for regulation (digging a moat) to ensure systems are "safe" and "aligned" while ignoring that humans are not aligned, so these systems obviously cannot be aligned with humans; they can only be aligned with their owners (i.e. them, not you).

Alignment in the realm of AGI is not about getting everyone to agree. It's about whether or not the AGI is aligned to the goal you've given it. The paperclip AGI example is often used, you tell the AGI "Optimize the production of paperclips" and the AGI started blending people to extract iron from their blood to produce more paperclips. Humans are used to ordering around other humans who would bring common sense and…

Right and that's why it's a farce.

> Whoa whoa whoa, we can't let just anyone run these models. Only large corporations who will use them to addict children to their phones and give them eating disorders and suicidal ideation, while radicalizing adults and tearing apart society using the vast profiles they've collected on everyone through their global panopticon, all in the name of making people unhappy so that it's easier to sell them more crap they don't need (a goal which is itself a problem in the face of an impending climate crisis). After all, we wouldn't want it to end up harming humanity by using its superior capabilities to manipulate humans into doing things for it to optimize for goals that no one wants!

Post reply on HN