Be skeptical of OpenAI’s rogue hacker agent story | John Thickstun

If OpenAI loudly proclaims how dangerous AI is, investors will hear how powerful it is. And who benefits from that? On 14 February 2019, OpenAI announced a language model called GPT-2, the precursor to the models that power modern AI chatbots and agents such as ChatGPT and Claude. But OpenAI declared GPT-2 was too risky to release, citing concerns about safety and abuse. I recall being annoyed at the time that OpenAI would make such a useless announcement: the risks seemed overblown, and without access to the model there wasn’t much for a researcher like me to learn about GPT-2. Continue reading...

John Thickstun · Fri, Jul 24, 2026, 6:00 AM
US news | The Guardian

Be skeptical of OpenAI’s rogue hacker agent story | John Thickstun

If OpenAI loudly proclaims how dangerous AI is, investors will hear how powerful it is. And who benefits from that?

On 14 February 2019, OpenAI announced a language model called GPT-2, the precursor to the models that power modern AI chatbots and agents such as ChatGPT and Claude. But OpenAI declared GPT-2 was too risky to release, citing concerns about safety and abuse.

I recall being annoyed at the time that OpenAI would make such a useless announcement: the risks seemed overblown, and without access to the model there wasn’t much for a researcher like me to learn about GPT-2.

This page shows an excerpt; reporting belongs to the original publisher. Some images or embeds may be omitted compared with the live article.