So what's the business strategy here?

Google is the only USA based frontier lab releasing open models. I know they aren't doing it out of the goodness of their hearts.

▲

artificialprint 2 days ago | parent | next [-]

Release open weights so competitors can't raise good money, then rear naked choke when they run dry

▲

robocat a day ago | parent [-]

Using Brazilian Jiu-Jitsu (BJJ) technical terms is confusing. Sports allusions don't travel well between cultures, especially if they sound seedy.

	▲	golfer a day ago \| parent [-]
		I found it plucky and intriguing. A great metaphor, not often seen in tech. Not everything has to be in the lowest common denominator of language.

▲

g947o 2 days ago | parent | prev | next [-]

https://openai.com/index/introducing-gpt-oss/

▲

stavros a day ago | parent [-]

This is nearly a year old, which is a million years in LLM time.

▲

g947o a day ago | parent [-]

*8 months

That doesn't make parent's claim true or even relevant.

And OpenAI could release an open model tomorrow. Nobody knows.

▲

stavros a day ago | parent [-]

Anybody could release an open model tomorrow. Google is the only US based lab releasing open weights models. OpenAI released one once, which might or might not count as "releasing", depending on your definition

	▲	g947o a day ago \| parent \| next [-]
		> OpenAI released one once You forgot the GPT-2 that came long before that. OpenAI was the lab that releases open models. None of this is factually correct, that is it. I don't think this is debatable. I don't love OpenAI, but OpenAI made huge contributions to the field, and one should give credit where credit is due. I have great trouble understanding why someone would waste time defending it.
	▲	petu a day ago \| parent \| prev \| next [-]
		> OpenAI released one once they've released gpt-oss-safeguard in October I hope / think they are going to release more, just going for one big release a year like Gemma (if we talk strictly about general chat model -- Gemma 3 was March 2025)
	▲	nickthegreek 19 hours ago \| parent \| prev [-]
		> OpenAI released one once gpt-2, CLIP, Whisper, Point-E, got-oss-120b, gpt-oss-20b.

▲

BoingBoomTschak 12 hours ago | parent | prev [-]

Spreading propaganda through aligned model censored to eschew wrongthink? I mean, I truly believe there's some of that in the LLM world, but probably not the real reason you're searching for. Might be trying to (re)gain mindshare/cred amongst the hackers.