Live data from Hacker News

OpenAI drops ban on military tools to partner with The Pentagon

semafor.com

401–410 of 509 posts

Re: OpenAI drops ban on military tools to partner with The Pentagon

#401

Earlier quoted context omitted.

I sure hope so, Russia has 100% been doing it, with multiple documented „botfarms” (more like fake identity verification farms)

I've noticed a huge surge in negativity and pessimism on English-language social media within the last year or so, roughly corresponding with to the spread of LLM tech. I do wonder whether these people are mostly just bots.

Twitter and Reddit seem to be filled with bots and shills nowadays.

I think it is a consequence of actual users leaving for private groups (Discord, Telegram, Whatsapp) or small forums.

But ye surely LLMs are increasing the bot count.

Re: OpenAI drops ban on military tools to partner with The Pentagon

#402

Earlier quoted context omitted.

I wonder how the OpenAI employees feel about this after their full throated support for reinstating Altman and the shuffling of the previous board.

I've noticed that, in my circles at least, after Russia's invasion of Ukraine working in defence is no longer considered inherently immoral. Maybe they don't see anything wrong with it either.

I don't think people change that fast. The main difference is that being anti-war seems to have come to be associated with a social cost, I think.

Re: OpenAI drops ban on military tools to partner with The Pentagon

#403
post #207
post #198

Ok so a few weeks ago, virtually every OpenAi employee signed a letter saying “reinstate Altman or I quit”. Is any OpenAi employee wanting to sign a “don’t use OpenAi for war or I quit”? Or was everyone in on betraying all OpenAi’s founding principles from the start?

Reinstating Altman was all about the money, and so is this.

You’re assuming these are separate events.

Re: OpenAI drops ban on military tools to partner with The Pentagon

#404

Earlier quoted context omitted.

I think you are pretty close. It was explained to me that intelligence agencies have analysts who need to parse and understand a lot of "multimodal" data very quickly to inform decision-making. Maybe the thinking is that LLMs could help with it? I don't quite know how they would deal with hallucinations and uncertainties but possibly LLMs could do low-level work and the analyst would double-check it and provide their…

Take a look at what you can find about Google’s Project Nimbus. It’s my understanding from public info that this system is active in Israel today, and from day one of the war was doing multimodal analysis of the content posted online by Hamas to locate the hostages in some automated fashion.

Found it — very interesting. Thanks for the pointer!

Re: OpenAI drops ban on military tools to partner with The Pentagon

#405

The second OpenAI stopped being a not for profit it opened itself up to the highest bidder namely the military industrial complex ... this danger motivated it's corporate structure which explicitly detailed it did not wish to be a pawn of big corporations so sad

It just shows the ‘not for profit’ was just a ploy to sound cool. Really, it should have been ‘only for profit’.

And it reopens the discussion about all its training on unattributed scrapped data.

Re: OpenAI drops ban on military tools to partner with The Pentagon

#406

Earlier quoted context omitted.

OpenAI had/has 57 Chinese talent [1], they're probably getting the boot if OpenAI picks up Pentagon contracts. https://macropolo2.wpenginepowered.com/wp-content/uploads/20...

That would be a great loss of talent. I wish we could make it much easier for those top scientists to become American citizens. Why assume they have loyalty to the CCP? They’re in California, presumably they like hot dogs, monster trucks, and bikinis.

> Why assume they have loyalty to the CCP?

You answered

> They’re in California

Re: OpenAI drops ban on military tools to partner with The Pentagon

#407
post #207

Earlier quoted context omitted.

Reinstating Altman was all about the money, and so is this.

Or they received a classified briefing showing how far ahead China or other adversaries are in terms of AI for military use and felt obligated to not let the US fall further behind.

Last I heard the Chinese had found a way of fueling their rockets with tap water. We're screwed regardless of their AI prowess.

Re: OpenAI drops ban on military tools to partner with The Pentagon

#409

Earlier quoted context omitted.

You talk about a country defending itself as if that's what militaries do. The US military has been intervening in the world stage for 80 years at this point. Take any country below the equator and chances are the US has probably used its military against it. That's not defence. You don't collaborate with that actor if you think defence is the goal.

Not Australia, New Zealand, or South Africa!

Well England did nuke Australia. But we’re friends so thats OK.

If we set that aside as an aberration, then yes your right.

Re: OpenAI drops ban on military tools to partner with The Pentagon

#410

I'm confused - what exactly does the Pentagon have to gain from OpenAI's products? Their front-runners are an LLM and an image generator, and I'm not sure what large-scale use case the US Department of Defense can have for either of these.

ChatGPT is generally useful as a tool for most people.

If you work on any secret project, you're not supposed to just log into openai.com and ask ChatGPT about things related to the secret stuff.

That's why OpenAI/Microsoft has infrastructure to run separate instances of the models in an environment where they don't retain the chat logs and presumably have security audits etc. When the original military ban was in place, organizations in the Pentagon could not negotiate a contract with OpenAI for these deals.

I have no idea whether they have models that are oriented for military use specifically, but even the plain ChatGPT is likely useful for a variety of general tasks.

Post reply on HN