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

OpenAI says its new model GPT-2 is too dangerous to release (2019)

slate.com

Post reply on HN
Print view View original

111–120 of 131 posts

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#111

Post by 0rbiter »

raincole wrote: post #22
cinkhangin wrote: post #6

I think they are right unintentionally. The growing amount of low-quality content everywhere could become a real problem.

They were more than right. They were correct in an intentional, precise manner. This is what OpenAI actually stated[0]: > Synthetic imagery, audio, and video, imply that technologies are reducing the cost of generating fake content and waging disinformation campaigns. > ‘The public at large will need to become more sceptical of text they find online, just as the ”deep fakes” phenomenon calls for more scepticism about…

[dead]

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#112

Post by yaroslavvb »

adt wrote: post #98
yaroslavvb wrote: post #85

This was around the time I trained Transformer-XL (outside of OpenAI) with Ben Mann ( https://yaroslavvb.medium.com/scaling-transformer-xl-to-128-... ) . Originally we wanted to release train and release the weights as a kind of GPT-2.5, but our OpenAI friends pushed us to keep weights closed.

Same story with Connor Leahy and his GPT-2 clone, though his public articulation of how OpenAI sat him down seems to be glossed over. " OpenAI reached out to me almost immediately to talk and they were nothing but respectful and understanding... After making it publicly known what I had done, I was quickly approached by a range of smart people with good arguments. Many of them helped me update my beliefs in light of…

Thanks for the share! Didn't realize eleutherai launched around same time

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#113

Post by yaroslavvb »

blitzar wrote: post #102
yaroslavvb wrote: post #85

This was around the time I trained Transformer-XL (outside of OpenAI) with Ben Mann ( https://yaroslavvb.medium.com/scaling-transformer-xl-to-128-... ) . Originally we wanted to release train and release the weights as a kind of GPT-2.5, but our OpenAI friends pushed us to keep weights closed.

> our OpenAI friends I would be taking a grudge against those "friends" to my grave.

I don't hold grudge, GPT-2 wasn't that great of a model, so releasing it would be more of a publicity value. But the blog post already had that purpose.

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#114

Post by yaroslavvb »

yaroslavvb wrote: post #113
blitzar wrote: post #102

Earlier quoted context omitted.

> our OpenAI friends I would be taking a grudge against those "friends" to my grave.

I don't hold grudge, GPT-2 wasn't that great of a model, so releasing it would be more of a publicity value. But the blog post already had that purpose.

This project gave me motivation to build the deep learning next token prediction integration for JetBrains because I was using PyCharm at the time. (Eventually, it wasn't continued because it was kind of expensive to host)

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#115

Post by ryandrake »

dd8601fn wrote: post #107
serf wrote: post #73

Earlier quoted context omitted.

>> it's really hard to sometimes break out of that loop and do manual fixes it's not just an erosion of skills, it can also break the whole LLM toolchain flow. Easy example: Put together some fairly complicated multi-facet program with an LLM. You'll eventually hit a bug that it needs to be coaxed into fixing. In the middle of this bug-fixing conversation go and ahead and fire an editor up and flip a true/false or ch…

When I have to pop in and solve a problem, I tell it I fixed it and what was wrong. Depending on the depth of its misunderstanding it could become a memory note or a readme update. I haven’t had any real issues with that approach.

It sucks that you have to do this.

I'm not yet at the point where I'm comfortable with just vibe coding slop and committing to source control. I'm always going in and correcting things the LLM does wrong, and it really sucks to have to keep a mental list of all the changes you made, just so you can tell your Eager Electronic Intern that you made them deliberately and to not undo them or agonize over them.

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#116

Post by ryandrake »

raincole wrote: post #22
cinkhangin wrote: post #6

I think they are right unintentionally. The growing amount of low-quality content everywhere could become a real problem.

They were more than right. They were correct in an intentional, precise manner. This is what OpenAI actually stated[0]: > Synthetic imagery, audio, and video, imply that technologies are reducing the cost of generating fake content and waging disinformation campaigns. > ‘The public at large will need to become more sceptical of text they find online, just as the ”deep fakes” phenomenon calls for more scepticism about…

Most of "the public" is already fooled by terrible, obvious Photoshops. They don't have a chance with AI-generated content.

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#117

Post by loremium »

jjcm wrote: post #69

This is obviously in response to Mythos, but I'll actually defend their statement at that time - they were right to take a pause. Think about how much things have changed in our industry since GPT-2 has dropped - it WAS that dangerous, not in itself, but because it was the first that really signaled a change in the field of play. GPT-2 was where the capabilities of these were really proven, up until that point it was…

Gates Law

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#118

Post by salad-tycoon »

measurablefunc wrote: post #8

I'm wondering when people are going to figure out the doom marketing playbook.

We know it, we’re just susceptible to it. Like not eating for an extended time, you know you’ll get hungry and then you do. There’s a very basic but powerful response to danger and a need for safety.

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#119

Post by renewiltord »

subroutine wrote: post #91
renewiltord wrote: post #35

Earlier quoted context omitted.

No. That’s not true. https://huggingface.co/openai/gpt-oss-120b Was released after.

I'm confused. From what I can tell GPT-2 was released in 2019... https://huggingface.co/openai-community/gpt2 https://github.com/openai/gpt-2/blob/master/model_card.md

It was not the last version they released open source. GPT-OSS-120B was released open source after it.

Re: OpenAI says its new model GPT-2 is too dangerous to release (2019)

#120

Post by johnfn »

SilverSlash wrote: post #75
jeswin wrote: post #67

Earlier quoted context omitted.

> a very straightforward and basic UI bug Show us the code, or an obfuscated snippet. A common challenge with coding-agent related posts is that the described experiences have no associated context, and readers have no way of knowing whether it's the model, the task, the company or even the developer. Nobody learns anything without context, including the poster.

That's hard to believe in my case. I tried a variety of prompts, 3 different frontier models, provided manual screenshot(s), the agent itself also took its own screenshots from tests during the course of debugging. Nothing worked. I have now fixed the bug manually after 15-20 minutes of playing around with a codebase where I don't know the language and didn't write a single line of code until now.

What's hard to believe? OP just asked what the bug was.
Post reply on HN