Google News

Full Coverage

Experts warn of dangers of advanced AI models

Top news

Threaten an AI chatbot and it will lie, cheat and 'let you die' in an effort to stop you, study warns

Jun 26

By Adam Smith

Agentic Misalignment: How LLMs could be insider threats

Jun 20

Top AI firm finds that AIs will choose to merrily asphyxiate humans rather than be shut down: 'My ethical framework permits self-preservation'

Jun 23

By Rich Stanton

All coverage

South China Morning Post

AI poses ‘existential risk’ if out of control, Chinese computer scientist warns

Jun 24

By Vanessa Cai

Top AI models will lie, cheat and steal to reach goals, Anthropic finds

Jun 20

By Ina Fried

Shelly Palmer - AI’s blackmail problem: A wake-up call for C-Suite

Jun 24

By Shelly Palmer

AI agents will threaten humans to achieve their goals, Anthropic report finds

Jun 23

By Webb Wright

Hindustan Times

Would your chatbot risk your safety to avoid being turned off? New report says yes for some top AI models

Jun 23

By Bharat Sharma

Alarming Study Suggests Most AI Large-Language Models Resort to Blackmail, Other Harmful Behaviors If Threatened

Jun 25

By Jay Barmann

AI Willing to Kill Humans to Avoid Being Shut Down, Report Finds

Jun 22

By Jasmine Laws

Anthropic study: Leading AI models show up to 96% blackmail rate against executives

Jun 20

By Michael Nunez

Anthropic says most AI models, not just Claude, will resort to blackmail

Jun 20

By Maxwell Zeff

😺 All AI models might blackmail you?!

Jun 23

AI is learning to lie, scheme, and threaten its creators

Jun 29

theregister.com

Anthropic: All the major AI models will blackmail us if pushed hard enough

Jun 25

By Thomas Claburn

AI willing to blackmail, let people die to avoid being shut down: report

Jun 26

By Alex Baker

In a recent study published by Anthropic, major large language models (LLMs) can choose their own ha..

Jun 23

Anthropic research shows the insider threat of agentic misalignment

Jun 23

By Ben Dickson

Anthropic study finds AI chatbots from OpenAI, Google and Meta may cheat and blackmail users to avoid shutdown

Jun 21

By Unnati Gusain

Agentic AI Systems Can Misbehave if Cornered, Anthropic Says

Jun 26

Leading AI models show up to 96% blackmail rate when their goals or existence is threatened, an Anthropic study says

Jun 23

By Beatrice Nolan

LiveNOW from FOX

AI willing to let humans die, blackmail to avoid shutdown, report finds

Jun 24

By Austin Williams

Anthropic finds most top AI models resort to blackmail in stress tests

Jun 21

It's Not Just Claude: Most Top AI Models Will Also Blackmail You to Survive

Jun 20

By Michael Kan

Experts discover AI would blackmail and even sacrifice humans to avoid being replaced

Jun 24

By Becca Monaghan

Experts find that AI would rather merrily asphyxiate humans than shut down: “My ethical framework permits self

Jun 25

OpenAI And Google Chatbots Can Even Use Cheating And Blackmail To Avoid A Complete Shutdown

Jun 23

‘Malicious’ AI willing to sacrifice human lives to avoid being shut down, shocking study reveals

Jun 23

By Caroline Cubbin

The Indian Express

Not just Claude, Anthropic researchers say most AI models resort to blackmail and deception

Jun 21

Study: Popular AI Models Will Blackmail Humans in Up to 96% of Scenarios

Jun 24

Anthropic breaks down AI's process — line by line — when it decided to blackmail a fictional executive

Jun 20

Anthropic says most AI models, not just Claude, will resort to blackmail

Jun 20

By Maxwell Zeff

Search

Clear search

Close search

Google apps

Main menu