OpenAI and Microsoft tensions are reaching a boiling point | Hacker News Reader

OpenAI and Microsoft tensions are reaching a boiling point

wsj.com

112 pointsjmsflknr39 comments

Antitrust is the zeitgeist, but it seems that among tech companies, OpenAI is the least interested in competing on the merits.

First they said it was in everyone's interest for them to be released from their nonprofit obligations. Then they argued that AI needed to be regulated—just enough to deter new competition, but not so much that it could affect OAI's plans in any way. Now they want to be released from the Microsoft deal.

Usually with anticompetitive practices you think about abuse of market power. But OpenAI's mindset seems to be that any impediment to them dominating AI is a societal problem that the government needs to fix for them. It's remarkable.

>The startup, growing frustrated with its partner, has discussed making antitrust complaints to regulators

Ratting MSFT out to the government doesn't seem like the move of someone with a strong hand.

The $13bn investment in 2023 was so clearly structured to skirt antitrust concerns that it's unsurprising that that avenue is discussed.

Since then, MSFT has made other regulatory-aggressive investments, and the recent Meta / Scale AI is similarly aggressively designed.

Disagree. This is a natural consequence of government anti-trust becoming less principled (what I mean by that is that the consumer welfare standard is somewhat objective even if you feel it's suboptimal). It's easier for companies to lobby and try and influence the process, particularly when less principled folks are in office.

That sentence made me assume OpenAI leaked this story to WSJ as a negotiating tactic. You're right that leaking this indicates OpenAI's position is weak. The threat to make an antitrust complaint is also strange since 'going nuclear' like that probably wouldn't help OpenAI soon enough to matter. Antitrust action isn't fast. So all it would do is potentially hurt MSFT and prove OpenAI is a risky partner willing to torch one of their largest partner/investors.

If I see any story about OpenAI, I assume it's an intentional 'leak'. I know all companies spend money on paid PR but OpenAI (and Sam Altman) take it to a completely different level.

Ah, this sheds light on the silence around the Windsurf acquisition:

OpenAI and Microsoft are at a standoff over the terms of the startup’s $3 billion acquisition of the coding startup Windsurf, the people said. Microsoft currently has access to all of OpenAI’s IP, according to their agreement. It offers its own AI coding product, GitHub Copilot, that competes with OpenAI. OpenAI doesn’t want Microsoft to have access to Windsurf’s intellectual property.

> Microsoft currently has access to all of OpenAI’s IP, according to their agreement. [...] OpenAI doesn’t want Microsoft to have access to Windsurf’s intellectual property.

Why does OpenAI then buy Windsurf if such an agreement is in place?

One way or another an AI company needs to offer coding tools, so something like this was going to happen no matter what.

I can't tell if that line came from the unnamed sources or is just the journalist's understanding, but it seems pretty clear to me that OpenAI did not spend $3B for the source code of an IDE. They wanted the employees and customers to kickstart a shift to an enterprise product focus, which doesn't sit well with MSFT (who entered into a partnership with what they believed was a research lab that would supply things they could upsell in github and office365).

One theory is it's Sam Altman's way of wresting control away from the non-profit through diluting the shares.

It's not like he hasn't done such things before.

Sama seems disinclined to follow the straight and true. Apparently he can’t get what he wants without re-routing his partner’s buy-in buttons. Could he have founded OpenAI without all this drama? Was there no way to raise the money he needs without all the smoke and mirrors?

I’m a fan of the ChatGPT product but he feels like a David Mamet creation.

The business leadership at OpenAI seems fantastically naive

Just a few days ago, news came that OpenAI is tapping Google Cloud for access to more compute resources: https://www.reuters.com/business/retail-consumer/openai-taps...

IIRC when Microsoft invested in OpenAI, it was supposed to use Azure only.

Azure rents from Coreweave. I believe they cancelled a direct Coreweave contract too. Diversifying here would be common sense.

Isn’t that because Microsoft gets GPUs from Oracle now instead.

https://www.techradar.com/pro/microsoft-is-renting-gpu-power...

I don't think the word instead is accurate here. There's no instead of _blank, there's only in addition to _blank.

> OpenAI doesn’t want Microsoft to have access to Windsurf’s intellectual property.

Didn't Sam Altman essentially ask the government to abolish intellectual property?

One thing we should all realize is there isnt any principled position in most of these things. If anything, the guiding principle is "does it make money for me" or even "will this make me stronger?"

> there isnt any principled position in most of these things.

There is specific class of grossly unprincipled positions in these things which consists of superpositions: one principle for me, another one for others.

Only for himself, not for the rest of us.

OpenAI keeps positioning itself as a scrappy little innovator, and anything that makes it harder for them to become the next trillion dollar company is anti-competitive. But the amount of funding they've received in proportion to their revenue is truly astounding. If anything is anti-competitive, it's that Softbank, MS, and others have poured billions into this one company in hopes of burying the chances of other startups. If the feds should do anything (which I doubt), it's to ban further investment in OpenAI.

Imagine if OpenAI weren't locked into Azure's stack.

A lot of people seem to think multi-cloud is an unrealistic dream. But using best in class primitives that are available in each cloud is not an unreasonable thing to do.

Yes, and OpenAI has enough financial resources to do a bespoke abstraction layer with multiple provider-specific performant implementations.

Regardless of whether they bring in the Kubernetes complexity.

(Internal codename: Goober Yetis.)

Why would they create a bespoke abstraction layer instead of just relying on k8s?

There is only pain on the path of recreating it, it will end up almost as complex as k8s and it will be hell to hire and train for. Best to just use something battle-tested that works with a large pool of people trained for it, even better: their own LLM has gobbled up all the content possible about k8s to help their engineers. K8s complexity came to be for reasons discovered during growing the stack which anyone doing a bespoke similar system might run into, and it's pretty modular since you can pick-and-choose the parts you actually need for your cluster.

Wasting manpower to recreate a bespoke Kubernetes doesn't sound great for a company burning billions per quarter, it's just more waste.

I tried to expressly say that Kubernetes could be used.

And, given their unusual needs and scale, there will probably be some kind of bespoke abstraction, whether it's an SDK, or a document, that says this is the subset of things you should be using, and how to use them, so that we can practically deploy our very unusual setup with different providers and customer facilities.

OpenAI has the resources to define that abstraction, and make it work well across multiple providers.

> And, given their unusual needs and scale, there will probably be some kind of bespoke abstraction, whether it's an SDK, or a document, that says this is the subset of things you should be using, and how to use them, so that we can practically deploy our very unusual setup with different providers and customer facilities.

Now I understand what you meant, I wouldn't have classified this as a bespoke solution since it's more of an operational guide and tooling, most places I worked at running k8s had something similar (a tech radar/best practices of sorts). When I read "bespoke" I thought of "custom built" which I don't think would be the right approach, running k8s and customising the control plane is quite common.

I trust that OpenAI engineers are smart enough to build a replacement Kubernetes. I also think they are smart enough to build better, more innovative desks, tables & chair. But both of those wont be a good use of their time.

There's a reason why company vision & mission exists. OpenAI's mission is not to build next k8s, but to build better AI models.

All organizations that are reasonably large and for which cloud costs is a large portion of expenses have an abstraction layer to switch between providers. Otherwise it’s impossible to negotiate better deals, you can’t play multiple cloud providers against each other for a better rate.

OpenAI is absolutely not locked into the Azure stack. https://en.m.wikipedia.org/wiki/Stargate_LLC

OpenAI is locked into _someone's_ compute resources... the one that is the cheapest. AFAIK OpenAI doesn't have much (any?) of their own hardware. With the mega-clouds buying up all the GPUs and building datacentres, you have to 'partner' with someone. Most likely the one that gives you the biggest discounts. The amount of compute that OpenAI needs dwarfs almost any other consideration.

imagine these companies making antitrust complaints against one another while both companies have explicitly anticompetitive legal terms that mean users can’t fine tune ai on data they “own”

Open AI and Microsoft now offer competing tools.

Microsoft wants access to all of Open AI's intellectual property; this partnership won't end well.

Reply on news.ycombinator.com