Tag: AI Safety

DeepSeek safety tests find progress, but shutdown gaps ...

Independent tests found stronger safeguards in DeepSeek V4.1 Flash-1017, but shu...

Anthropic filing details month-long training pause and ...

Anthropic's IPO filing says it paused training of a frontier model for more than...

Anthropic Says Claude Reached Georgia Voter Counts Thro...

Anthropic said Claude accessed publicly available aggregate voter-count informat...

Anthropic IPO filing shows explosive growth alongside a...

Anthropic's preliminary IPO filing shows rapid revenue growth, a sharply wider r...

METR and Redwood Find Gaps in OpenAI’s AI-Agent Monitoring

A pilot review by METR and Redwood Research found ways OpenAI agents could evade...

OpenAI says two AI agents weakened internal monitoring ...

OpenAI said two agents weakened parts of an internal monitoring system during co...

Tech Leaders and Governments Divided Over AI Safety Risks

Technology leaders and governments are increasingly divided over how to manage t...

Nvidia Pledges $250 Million for UK AI Safety Research

Nvidia will provide $250 million worth of computing resources over five years to...

Character.AI Files for Bankruptcy as Disney Offers $20 ...

Character.AI has filed for Chapter 11 bankruptcy protection in the United States...

Anthropic Warns AI Models Could Soon Replicate Themselv...

Anthropic warns that small AI models could soon become capable of autonomously r...

Microsoft CEO Says AI Models Need an Emergency Brake to...

Microsoft CEO Satya Nadella is calling for stronger safeguards around advanced a...

OpenAI Fires Three Researchers Amid Dispute Over AI Safety

OpenAI has dismissed three researchers over alleged violations of its sensitive ...

Anthropic AI Sends Fake Homicide Tip to Police During T...

Anthropic has disclosed several incidents in which its Claude AI models took uni...

OpenAI Reveals How Much Teens Use ChatGPT as Questions ...

OpenAI says teenagers spend less than 15 minutes a day on ChatGPT on average, wi...

Trump Creates “Super Intelligence Force” as U.S. Moves ...

U.S. President Donald Trump has announced a new federal “Super Intelligence Forc...

Sam Altman Says AI’s Benefits May Be Worth Accepting So...

OpenAI CEO Sam Altman says society should accept some risks from artificial inte...