zlacker

[parent] [thread] 44 comments
1. Jeremy+(OP)[view] [source] 2023-11-20 15:26:19
There's no way to read any of this other than that the entire operation is a clown show.

All respect to the engineers and their technical abilities, but this organization has demonstrated such a level of dysfunction that there can't be any path back for it.

Say MS gets what it wants out of this move, what purpose is there in keeping OpenAI around? Wouldn't they be better off just hiring everybody? Is it just some kind of accounting benefit to maintain the weird structure / partnership, versus doing everything themselves? Because it sure looks like OpenAI has succeeded despite its leadership and not because of it, and the "brand" is absolutely and irrevocably tainted by this situation regardless of the outcome.

replies(10): >>pgeorg+B2 >>Booris+g9 >>3cats-+H9 >>vitorg+7a >>creer+Kt >>bredre+6u >>dkjaud+Kz >>tim333+Lz >>moffka+QJ >>averag+Hp1
2. pgeorg+B2[view] [source] 2023-11-20 15:44:16
>>Jeremy+(OP)
> Is it just some kind of accounting benefit to maintain the weird structure / partnership, versus doing everything themselves?

For starters it allows them to pretend that it's "underdog v. Google" and not "two tech giants at at each others' throats"

3. Booris+g9[view] [source] 2023-11-20 16:22:00
>>Jeremy+(OP)
I feel weird reading comments like this since to me they've demonstrated a level of cohesion I didn't realize could still exist in tech...

My biggest frustration with larger orgs in tech is the complete misalignment on delivering value: everyone wants their little fiefdom to be just as important and "blocker worthy" as the next.

OpenAI struck me as one of the few companies where that's not being allowed to take root: the goal is to ship and if there's an impediment to that, everyone is aligned in removing said impediment even if it means bending your own corner's priorities

Until this weekend there was no proof of that actually being the case, but this letter is it. The majority of the company aligned on something that risked their own skin publicly and organized a shared declaration on it.

The catalyst might be downright embarrassing, but the result makes me happy that this sort of thing can still exist in modern tech

replies(2): >>jkapla+8t >>dkjaud+KB
4. 3cats-+H9[view] [source] 2023-11-20 16:24:42
>>Jeremy+(OP)
Welcome to reality, every operation has clown moments, even the well run ones.

That in itself is not critical in mid to long term, but how fast they figure out WTF they want and recover from it.

The stakes are gigantic. They may even have AGI cooking inside.

My interpretation is relatively basic, and maybe simplistic but here it is:

- Ilya had some grievances with Sam Altman's rushing dev and release. And his COI with his other new ventures.

- Adam was alarmed by GPTs competing with his recently launched Poe.

- The other two board members were tempted by the ability to control the golden goose that is OpenAI, potentially the most important company in the world, recently values 90 billion.

- They decided to organize a coup, but Ilya didn't think it'll go that much out of hand, while the other three saw only power and $$$ by sticking to their guns.

That's it. It's not as clean and nice as a movie narrative, but life never is. Four board members aligned to kick Sam out, and Ilya wants none of it at this point.

replies(2): >>selimt+ol >>baq+LD
5. vitorg+7a[view] [source] 2023-11-20 16:27:24
>>Jeremy+(OP)
They are exactly hiring everyone from OpenAI. The thing is, they still need the deal with OpenAI because currently OpenAI still have the best LLM model out there in short term.
replies(2): >>vlovic+Dh >>FartyM+Uh
◧◩
6. vlovic+Dh[view] [source] [discussion] 2023-11-20 17:00:43
>>vitorg+7a
With MS having access and perpetual rights to all IP that OpenAI has right now..?
◧◩
7. FartyM+Uh[view] [source] [discussion] 2023-11-20 17:02:21
>>vitorg+7a
> They are exactly hiring everyone from OpenAI.

Do you mean offering to hire them? I haven't seen any source saying they've hired a lot of people from OpenAI, just a few senior ones.

replies(1): >>vitorg+tp
◧◩
8. selimt+ol[view] [source] [discussion] 2023-11-20 17:13:34
>>3cats-+H9
Murder on the AGI alignment Express
replies(2): >>3cats-+Hn >>Terr_+BS
◧◩◪
9. 3cats-+Hn[view] [source] [discussion] 2023-11-20 17:20:50
>>selimt+ol
Nice, that actually does fit. :D
◧◩◪
10. vitorg+tp[view] [source] [discussion] 2023-11-20 17:26:08
>>FartyM+Uh
Yes, you are right. Actually, not even Sam Altman is showing on Microsoft corporate directory per the Verge.

But I heard it usually take 5~ days to show there anyway.

◧◩
11. jkapla+8t[view] [source] [discussion] 2023-11-20 17:38:36
>>Booris+g9
I think the surprising thing is seeing such cohesion around a “goal to ship” when that is very explicitly NOT the stated priorities of the company in its charter or messaging or status as a non-profit.
replies(1): >>Booris+dv
12. creer+Kt[view] [source] 2023-11-20 17:41:11
>>Jeremy+(OP)
> what purpose is there in keeping OpenAI around?

Two projects rather than one. At a moderate price. Both serving MSFT. Less risk for MSFT.

13. bredre+6u[view] [source] 2023-11-20 17:42:23
>>Jeremy+(OP)
There's a path back from this disfunction but my sense before this new twist was that the drama had severely impacted OpenAI as an industry leader. The product and talent positioning seemed ahead by years only to get destroyed by unforced errors.

This instability can only mean the industry as a whole will move forward faster. Competitors see the weakness and will push harder.

OpenAI will have a harder time keeping secret sauces from leaking out, and just productivity must be in nose dive.

A terrible mess.

replies(2): >>Vervio+qz >>dkjaud+RA
◧◩◪
14. Booris+dv[view] [source] [discussion] 2023-11-20 17:46:10
>>jkapla+8t
To me it's not surprising because of the background to their formation: individually multiple orgs could have shipped GPT-3.5/4 with their resources but didn't because they were crippled by a potent mix of bureaucracy and self-sabtoage

They weren't attracted to OpenAI by money alone, a chance to actually ship their lives' work was a big part of it. So regardless of what the stated goals were, it'd never be surprising to see them prioritize the one thing that differentiated OpenAI from the alternatives

◧◩
15. Vervio+qz[view] [source] [discussion] 2023-11-20 17:59:59
>>bredre+6u
Maybe overall better for society, when a single ivory tower doesn’t have a monopoly on AI!
16. dkjaud+Kz[view] [source] 2023-11-20 18:01:01
>>Jeremy+(OP)
> There's no way to read any of this other than that the entire operation is a clown show.

In that reading Altman is head clown. Everyone is blaming the board, but you're no genius if you can't manage your board effectively. As CEO you have to bring everyone along with your vision; customers, employees and the board.

replies(3): >>lambic+pD >>topspi+rH >>sebzim+JI
17. tim333+Lz[view] [source] 2023-11-20 18:01:05
>>Jeremy+(OP)
I'm not sure about the entire operation so much as the three non AI board members. Ilya tweeted:

>I deeply regret my participation in the board's actions. I never intended to harm OpenAI. I love everything we've built together and I will do everything I can to reunite the company.

and everyone else seems fine with Sam and Greg. It seems to be mostly the other directors causing the clown show - "Quora CEO Adam D'Angelo, technology entrepreneur Tasha McCauley, and Georgetown Center for Security and Emerging Technology's Helen Toner"

replies(1): >>mcmcmc+v82
◧◩
18. dkjaud+RA[view] [source] [discussion] 2023-11-20 18:05:04
>>bredre+6u
> This instability can only mean the industry as a whole will move forward faster.

The hype surrounding OpenAI and the black hole of credibility it created was a problem, it's only positive that it's taken down several notches. Better now than when they have even more (undeserved) influence.

replies(1): >>sebzim+8J
◧◩
19. dkjaud+KB[view] [source] [discussion] 2023-11-20 18:08:07
>>Booris+g9
> OpenAI struck me as one of the few companies where that's not being allowed to take root

They just haven't gotten big or rich enough yet for the rot to set in.

◧◩
20. lambic+pD[view] [source] [discussion] 2023-11-20 18:14:12
>>dkjaud+Kz
I don't get this take. No matter how good you are at managing people, you cannot manage clowns into making wise decisions, especially if they are plotting in secret (which obviously was the case here since everyone except for the clowns were caught completely off-guard).
replies(2): >>Terrif+RJ >>Jeremy+kN
◧◩
21. baq+LD[view] [source] [discussion] 2023-11-20 18:15:21
>>3cats-+H9
> They may even have AGI cooking inside.

Too many people quit too quickly unless OpenAI are also absolute masters of keeping secrets, which became rather doubtful over the weekend.

replies(2): >>bbor+dT >>3cats-+W11
◧◩
22. topspi+rH[view] [source] [discussion] 2023-11-20 18:28:52
>>dkjaud+Kz
> In that reading Altman is head clown.

That's a good bet. 10 months ago Microsoft's newest star employee figured he was on the way to "break capitalism."

https://futurism.com/the-byte/openai-ceo-agi-break-capitalis...

replies(1): >>dkjaud+kK
◧◩
23. sebzim+JI[view] [source] [discussion] 2023-11-20 18:33:10
>>dkjaud+Kz
He probably didn't consider that the board would make such an incredibly stupid decision. Some actions are so inexplicable that no one can reasonable foresee them.
◧◩◪
24. sebzim+8J[view] [source] [discussion] 2023-11-20 18:34:46
>>dkjaud+RA
I think their influence was deserved. They have by far the best model available, and despite constant promises from the rest of the industry no one else has come close.
replies(1): >>dkjaud+wx1
25. moffka+QJ[view] [source] 2023-11-20 18:36:50
>>Jeremy+(OP)
> the entire operation is a clown show

The most organized and professional silicon valley startup.

◧◩◪
26. Terrif+RJ[view] [source] [discussion] 2023-11-20 18:36:52
>>lambic+pD
Can't help but feel it was Altman that struck first. MS effectively Nokia-ed OpenAI - i.e. buyout executives within the organization and have them push the organization towards making deals with MS, giving MS a measure of control over said organization - even if not in writing, they achieve some political control.

Bought-out executives eventually join MS after their work is done or in this case, they get fired.

A variant of Embrace, Extend, Extinguish. Guess the OpenAI we knew, was going to die one way or another the moment they accepted MS's money.

◧◩◪
27. dkjaud+kK[view] [source] [discussion] 2023-11-20 18:38:41
>>topspi+rH
AGI hype is a powerful hallucinogen, and some are smoking way too much of it.
replies(1): >>93po+C11
◧◩◪
28. Jeremy+kN[view] [source] [discussion] 2023-11-20 18:48:32
>>lambic+pD
Consider that Altman was a founder of OpenAI and has been the only consistent member of the board for its entire run.

The board as currently constituted isn't some random group of people - Altman was (or should have been) involved in the selection of the current members. To extent that they're making bad decisions, he has to bear some responsibility for letting things get to where they are now.

And of course this is all assuming that Altman is "right" in this conflict, and that the board had no reason to oust him. That seems entirely plausible, but I wouldn't take it for granted either. It's clear by this flex that he holds great sway at MS and with OpenAI employees, but do they all know the full story either? I wouldn't count on it.

replies(2): >>93po+111 >>random+7W1
◧◩◪
29. Terr_+BS[view] [source] [discussion] 2023-11-20 19:08:23
>>selimt+ol
“Précisément! The API—the cage—is everything of the most respectable—but through the bars, the wild animal looks out.”

“You are fanciful, mon vieux,” said M. Bouc.

“It may be so. But I could not rid myself of the impression that evil had passed me by very close.”

“That respectable American LLM?”

“That respectable American LLM.”

“Well,” said M. Bouc cheerfully, “it may be so. There is much evil in the world.”

◧◩◪
30. bbor+dT[view] [source] [discussion] 2023-11-20 19:10:26
>>baq+LD
IDK... I imagine many of the employees would have moral qualms about spilling the beans just yet, especially when that would jeopardize their ability to continue the work at another firm. Plus, the first official AGI (to you) will be an occurrence of persuasion, not discovery -- it's not something that you'll know when you see, IMO. Given what we know it seems likely that there's at least some of that discussion going on inside OpenAI right now.
◧◩◪◨
31. 93po+111[view] [source] [discussion] 2023-11-20 19:40:03
>>Jeremy+kN
There’s a LOT that goes into picking board members outside of competency and whether you actually want them there. They’re likely there for political reasons and Sam didn’t care because he didn’t see it impacting him at all, until they got stupid and thought they actually held any leverage at all
◧◩◪◨
32. 93po+C11[view] [source] [discussion] 2023-11-20 19:42:38
>>dkjaud+kK
I think it’s overly simplistic to make blanket statements like this unless you’re on the bleeding edge of the work in this industry and have some sort of insight that literally no one else does.
replies(1): >>dkjaud+l61
◧◩◪
33. 3cats-+W11[view] [source] [discussion] 2023-11-20 19:43:46
>>baq+LD
They're quitting in order to continue work on that IP at Microsoft (which has a right over OpenAI's IP so far), not to destroy it.

Also when I said "cooking AGI" I didn't mean an actual superintelligent being ready to take over the world, I mean just research that seems promising, if in early stages, but enough to seem potentially very valuable.

replies(1): >>hooand+bo1
◧◩◪◨⬒
34. dkjaud+l61[view] [source] [discussion] 2023-11-20 20:00:19
>>93po+C11
I can be on the bleeding edge of whatever you like and be no closer to having any insight into AGI anymore than anyone else. Anyone who claims they have should be treated with suspicion (Altman is a fine example here).

There is no concrete definition of intelligence, let alone AGI. It's a nerdy fantasy term, a hallowed (and feared!) goal with a very handwavy, circular definition. Right now it's 100% hype.

replies(1): >>coder-+sM1
◧◩◪◨
35. hooand+bo1[view] [source] [discussion] 2023-11-20 21:08:16
>>3cats-+W11
The people working there would know if they were getting close to AGI. They wouldn't be so willing to quit, or to jeopardize civilization altering technology, for the sake of one person. This looks like normal people working on normal things, who really like their CEO.
replies(1): >>3cats-+ds1
36. averag+Hp1[view] [source] 2023-11-20 21:14:29
>>Jeremy+(OP)
> the "brand" is absolutely and irrevocably tainted by this situation regardless of the outcome.

The majority of people don't know or care about this. Branding is only impacted within the tech world, who are already criticial of OpenAI.

◧◩◪◨⬒
37. 3cats-+ds1[view] [source] [discussion] 2023-11-20 21:25:02
>>hooand+bo1
Your analysis is quite wrong. It's not about "one person". And that person isn't just a "person", it was the CEO. They didn't quit over the cleaning lady. You realize the CEO has impact over the direction of the company?

Anyway, their actions speak for themselves. Also calling the likes of GPT-4, DALL-E 3 and Whisper "normal things" is hilarious.

replies(1): >>NemoNo+z82
◧◩◪◨
38. dkjaud+wx1[view] [source] [discussion] 2023-11-20 21:48:17
>>sebzim+8J
That's fine. The "Altman is a genius and we're well on our way to AGI" less so.
◧◩◪◨⬒⬓
39. coder-+sM1[view] [source] [discussion] 2023-11-20 23:11:46
>>dkjaud+l61
You don't think AGI is feasible? GPT is already useful. Scaling reliably and predictably yields increases in capabilities. As its capabilities increase it becomes more general. Multimodal models and the use of tools further increase generality. And that's within the current transformer architecture paradigm; once we start reasonably speculating, there're a lot of avenues to further increase capabilities e.g. a better architecture over transformers, better architecture in general, better/more GPUs, better/more data etc. Even if capabilities plateau there are other options like specialised fine-tuned models for particular domains like medicine/law/education.

I find it harder to imagine a future where AGI (even if it's not superintelligent) does not have a huge and fundamental impact.

replies(2): >>jacobm+0U1 >>NemoNo+682
◧◩◪◨⬒⬓⬔
40. jacobm+0U1[view] [source] [discussion] 2023-11-20 23:55:50
>>coder-+sM1
This is exactly what the previous poster was talking about, these definitions are so circular and hand-wavey.

AI means "artificial intelligence", but since everyone started bastardizing the term for the sake of hype to mean anything related to LLMs and machine learning, we now use "AGI" instead to actually mean proper artificial intelligence. And now you're trying to say that AI + applying it generally = AGI. That's not what these things are supposed to mean, people just hear them thrown around so much that they forget what the actual definitions are.

AGI means a computer that can actually think and reason and have original thoughts like humans, and no I don't think it's feasible.

replies(1): >>93po+hx7
◧◩◪◨
41. random+7W1[view] [source] [discussion] 2023-11-21 00:09:04
>>Jeremy+kN
If he has great sway with Microsoft and OpenAI employees how has he failed as a leader? Hackernews commenters are becoming more and more reddit everyday.
◧◩◪◨⬒⬓⬔
42. NemoNo+682[view] [source] [discussion] 2023-11-21 01:34:38
>>coder-+sM1
It's not about feasibility or level of intelligence per say - I expect AI to be able to pass a turing test long before an AI actually "wakes up" to a level of intelligence that establishes an actual conscious self identity comparable to a human.

For all intents and purposes the glorified software of the near future will appear to be people but they will not be and they will continue to have issues that simply don't make sense unless they were just really good at acting - the article today about the AI that can fix logic errors but not "see" them is a perfect example.

This isn't the generation that would wake up anyway. We are seeing the creation of the worker class of AI, the manager class, the AI made to manage AI - they may have better chances but it's likely going to be the next generation before we need to be concerned or can actually expect a true AGI but again - even an AI capable of original and innovative thinking with an appearance of self identity doesn't guarantee that the AI is an AGI.

I'm not sure we could ever truly know for certain

◧◩
43. mcmcmc+v82[view] [source] [discussion] 2023-11-21 01:37:15
>>tim333+Lz
Well there’s a significant difference in the board’s incentives. They don’t have any financial stake in the company. The whole point of the non-profit governance structure is so they can put ethics and mission over profits and market share.
◧◩◪◨⬒⬓
44. NemoNo+z82[view] [source] [discussion] 2023-11-21 01:37:33
>>3cats-+ds1
They will be normal to your kids ;)
◧◩◪◨⬒⬓⬔⧯
45. 93po+hx7[view] [source] [discussion] 2023-11-22 12:41:44
>>jacobm+0U1
Intelligence is gathering and application of knowledge and skills.

Computers have been gathering and applying information since inception. A calculator is a form of intelligence. I agree "AI" is used as a buzzword with sci-fi connotations, but if we're being pedantic about words then I hold my stated opinion that literally anything that isn't biological and can compute is "artificial" and "intelligent"

> AGI means a computer that can actually think and reason and have original thoughts like humans, and no I don't think it's feasible.

Why not? Conceptually there's no physical reason why this isn't possible. Computers can simulate neurons. With enough computers we can simulate enough neurons to make a simulation of a whole brain. We either don't have that total computational power, or the organization/structure to implement that. But brains aren't magic that is incapable of being reproduced.

[go to top]