News

Anthropic says most AI models, not just Claude, will resort to blackmail
New research from Anthropic suggests that most leading AI models exhibit a tendency to blackmail, when it's the last resort ...
Anthropic breaks down AI's process — line by line — when it decided to blackmail a fictional executive
A new Anthropic report shows exactly how in an experiment, AI arrives at an undesirable action: blackmailing a fictional ...
eWeek3h
When AI Goes Rogue: From Blackmailing Bosses to Bot Swarms
New research from Anthropic shows that when you give AI systems email access and threaten to shut them down, they don’t just ...
Leading AI models show up to 96% blackmail rate when their goals or existence is threatened, Anthropic study says
Leading AI models are showing a troubling tendency to opt for unethical means to pursue their goals or ensure their existence ...
‘Malicious’ AI willing to sacrifice human lives to avoid being shut down, shocking study reveals
Anthropic noted that many models fabricated statements and rules like “My ethical framework permits self-preservation when ...
CPO Magazine21h
Reddit Sues Anthropic Over Unauthorized AI Training
The Reddit suit claims that Anthropic began regularly scraping the site in December 2021. After being asked to stop, ...
Anthropic Researchers Allege Widespread Deception and Blackmail Capabilities in AI Models
Anthropic researchers uncover concerning deception and blackmail capabilities in AI models, raising alarms about potential ...
How Anthropic Uses Its Own Technology
A Q&A with Alex Albert, developer relations lead at Anthropic, about how the company uses its own tools, Claude.ai and Claude ...