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

Corporate America is getting hooked on open-source AI

nytimes.com

Post reply on HN
Print view View original

121–130 of 303 posts

Re: Corporate America is getting hooked on open-source AI

#121

Post by hungryhobbit »

cmiles8 wrote: post #71
euroderf wrote: post #63

Earlier quoted context omitted.

So the business challenge is to balance sizable investments and relatively small marginal costs. Not so different from other digital goods.

Fair assessment. The challenge for OpenAI and Anthropic is that they need sizeable margins to pay for the massive costs incurred. Market forces are driving things in the opposite direction and fast. When your competition has a tiny cost base compared to yours and lacks the bonkers future capital commits you made then that’s a terrible position to be in… hence their conundrum.

Even "massive costs" is an understatement: these companies have astronomically massive costs!

Take Open AI for instance: it has "zero debt" ... and $665 billion to $1.4 trillion in "long-term commitments".

Re: Corporate America is getting hooked on open-source AI

#122

Post by techpression »

senordevnyc wrote: post #117
techpression wrote: post #76

Earlier quoted context omitted.

If you use OpenAI or Anthropic trust were never a requirement in the first place.

There’s zero evidence that OpenAI or Anthropic has ever violated their enterprise agreements and trained on data they agreed not to, and that’s what these big companies care about. They don’t care that you hate Sam and Dario.

There’s more to trust than training on data? Most of their products are vibe coded, bug ridden and their uptime is some of the worst in the history of services. How many IT departments let cowork run around freely?

Re: Corporate America is getting hooked on open-source AI

#124

Post by jhonof »

r_lee wrote: post #47
syntaxing wrote: post #22

I swear, Qwen 3.8 27B @ Q8 is smarter than Sonnet 5 most of the time. Why wouldn’t corporate America self host at this point, especially with better options like Deepseek Flash and GLM 5.3 flash that’s a middle ground between Sonnet and Opus

is this actually the case? I haven't kept up with the small models but if there's roughly Sonnet 4.6 level capable open small models, then I'd be impressed

Qwen 3.8 27B is better than Sonnet 4.6

Re: Corporate America is getting hooked on open-source AI

#125

Post by biophysboy »

hintymad wrote: post #81

I think using open-source AI is no longer about API cost but about company survival. Take Anthropic for an example. Anthropic has successfully destroyed customer trust, at least for me. DHH in a recent interview mentioned that Claude refused to translate an article about immigration. Not summarize. Not editorialize. Translate! I think this reveals an unacceptable level of paternalism: Anthropic fundamentally believes…

I like the Claude constitution overall - I hope it becomes something representatives vote on and amend, to avoid the centralized corporate censorship you describe. In the meantime, I am fine with it abstaining from doing DHH’s bidding, especially because there are so many AI alternatives.

Re: Corporate America is getting hooked on open-source AI

#126

Post by calebkaiser »

r_lee wrote: post #39
cmiles8 wrote: post #26

Every larger company I talk to these days has an active project on moving away from OpenAI and Anthropic to open models. And they’re actively shifting, as the article says, so the threat is far from theoretical. Unless they both dramatically slash prices then they’re in big trouble. Neither of them can afford to do that and both desperately need to convince the street that the opposite will happen if they want any ho…

I knew that there was no real moat from the very start, I mean, these things were close enough from the very start, how could it not result in a race to the bottom, especially as you can't really prevent distillation reliably?

I support and use open models as much as possible, but I'm not totally convinced that OAI or Anthropic have no moat, even as open models catch up to the frontier. Serving and inference are still hard problems when you're talking about a 2 trillion parameter model. Fine-tuning, if that remains a realistic need for businesses, is also a difficult infra problem at that scale. In the most bearish case, where there is no competitive advantage to using their models, big labs still have an advantage in this area.

Maybe there is some threshold where the price/quality math for your standard business tips in favor of smaller models and self-hosting the entire stack. I'd certainly love that.

Re: Corporate America is getting hooked on open-source AI

#127

Post by senordevnyc »

techpression wrote: post #122
senordevnyc wrote: post #117

Earlier quoted context omitted.

There’s zero evidence that OpenAI or Anthropic has ever violated their enterprise agreements and trained on data they agreed not to, and that’s what these big companies care about. They don’t care that you hate Sam and Dario.

There’s more to trust than training on data? Most of their products are vibe coded, bug ridden and their uptime is some of the worst in the history of services. How many IT departments let cowork run around freely?

Then why are these companies seeing unprecedented revenue growth? Who is spending these tens of billions of dollars? Maybe it won’t last, but it’s already happening right now, driven largely by corporate America. So yeah, maybe the orgs don’t trust the big labs in some vague sense, and yet they’re shelling out vast sums to them.

Re: Corporate America is getting hooked on open-source AI

#128

Post by jimbokun »

thatmf wrote: post #91
hintymad wrote: post #81

I think using open-source AI is no longer about API cost but about company survival. Take Anthropic for an example. Anthropic has successfully destroyed customer trust, at least for me. DHH in a recent interview mentioned that Claude refused to translate an article about immigration. Not summarize. Not editorialize. Translate! I think this reveals an unacceptable level of paternalism: Anthropic fundamentally believes…

Ah, yes, I'm sure the article that moral paragon DHH wished to translate was not at all harmful, and that this was a good-faith effort on his part /s While I agree that Claude can be overly paternalistic at times, how should it respond to a request to translate, say, bomb-making instructions? It's reasonable to me that it might refuse this.

How do you know what was in the article?

Also, you see zero distinction between hearing opinions on political topics you might find objectionable, and building a bomb to kill people?

Re: Corporate America is getting hooked on open-source AI

#129

Post by aff-vasileva »

cmiles8 wrote: post #79
gmadsen wrote: post #78

Earlier quoted context omitted.

The key difference is that the models upgrade multiple times a year. It is an inherently different than a commodity market

Yes, but changing models, even across providers, takes about two seconds and one line of code. It’s literally the least stickiest thing in the history of tech. Which is a big problem for these companies.

The API switch takes one line of code. The enterprise switch takes one line of code and twelve departments.

Re: Corporate America is getting hooked on open-source AI

#130

Post by biophysboy »

mylies43 wrote: post #101
woah wrote: post #85

Earlier quoted context omitted.

Really? It's worth it to you to spend 10 minutes thinking about how to prompt a dumber model to save $0.05? (not that open source models are dumber any more)

Well its def not $0.05, I just started using claude sonnet 5 and Ive found most simple questions might be 0.05 cents, a unit test is something like 0.10 -> 0.20 and small features and classes get into the individual dollars. Sure its a lot faster but at the end of the day its not cheap. Plus there is something to say about being in the drivers seat, youll have a much better idea of how it works instead of needing to…

Yes, they can’t magically resolve ambiguities in the prompt. Hence it is better to make something clear at the outset
Post reply on HN