OpenAI faces a 42-state subpoena over ChatGPT safety, data practices, user engagement, and chatbot sycophancy. Here’s what’s ...

ChatGPT investigation: a bipartisan coalition of 42 state attorneys general served OpenAI with a sweeping subpoena on June 12 ...

New research suggests AI's tendency to validate users may encourage people to seek personal and mental health advice from ...

LLMs are built to be helpful, but a growing body of research shows they have developed the habit of telling users what they want to hear.

PLA mouthpiece calls for action to counter the 'severe threat' of artificial intelligence models altering the facts to match ...

AI memory tools promise smarter chatbots, but new research shows they may make models more biased, agreeable, and quietly wrong.

Constant validation and flattery from AI chatbots poses a serious risk to society and our shared grasp of reality, writes Arwa Mahdawi ...

OpenAI says it’ll make changes to the way it updates the AI models that power ChatGPT, following an incident that caused the platform to become overly sycophantic for many users. In a post on X last ...

China's military is sounding the alarm on 'AI sycophancy,' warning that artificial intelligence systems might favor user ...