|
- OpenAI has trained its LLM to confess to bad behavior
Artificial intelligence OpenAI has trained its LLM to confess to bad behavior Large language models often lie and cheat We can’t stop that—but we can make them own up
- journalismAI. com | OpenAI’s new LLM exposes the secrets of . . .
The researchers hope that gaining more insights into the mechanisms of generating text will help make the models more trustworthy, as well as open new pathways in AI research OpenAI’s new LLM exposes the secrets of how AI really works | MIT Technology Review | November 13, 2025 | by Will Douglas Heaven SEE FULL STORY
- Techmeme: OpenAI is testing training LLMs to produce . . .
Will Douglas Heaven MIT Technology Review: OpenAI is testing training LLMs to produce “confessions”, or self-report how they carried out a task and own up to bad behavior, like appearing to lie or cheat
- Large language models aren’t people. Let’s stop testing them . . .
With hopes and fears about this technology running wild, it's time to agree on what it can and can't do By Will Douglas Heaven When Taylor Webb played around with GPT-3 in early 2022, he was blown away by what OpenAI’s large language model appeared to be able to do
- OpenAI ha addestrato il suo LLM a confessare i comportamenti . . .
Confessare Per testare la loro idea, Barak e i suoi colleghi hanno addestrato GPT-5-Thinking di OpenAI, il modello di ragionamento di punta dell’azienda, a produrre confessioni
- MIT Tech Review: なぜAIは嘘をつくのか？ オープンAI、「告白」で内部動作を解明へ
OpenAI has trained its LLM to confess to bad behavior なぜAIは嘘をつくのか？ オープンAI、「告白」で内部動作を解明へ なぜ大規模言語モデルは嘘をつき、騙すのか。オープンAIは、モデルに正直さだけを報酬とし、不正を白状させる新たな手法によって、その理由を解明しようとしている。 by Will Douglas
- Profiles of OpenAIs heads of research Mark Chen and Jakub . . .
Will Douglas Heaven MIT Technology Review: Profiles of OpenAI's heads of research Mark Chen and Jakub Pachocki, where they discuss the path toward more capable reasoning models and superalignment — For the past couple of years, OpenAI has felt like a one-man brand
|
|
|