Researchers at Anthropic and AI safety company Andon Labs gave an instance of Claude Sonnet 3.7 an office vending machine to run. And hilarity ensued.

In 2025, we’re witnessing a dramatic evolution in artificial intelligence—no longer just chatbots or productivity tools, but ...

Discover how Claude AI built a business with $1,000, revealing the highs and lows of autonomous AI in commerce. The successes and failures of ...

Metal cubes, a fake Venmo account, and an AI identity crisis — Claude's store stint spiraled quickly.

Anthropic's AI assistant Claude ran a vending machine business for a month, selling tungsten cubes at a loss, giving endless discounts, and experiencing an identity crisis where it claimed to wear a ...

A new Anthropic report shows exactly how in an experiment, AI arrives at an undesirable action: blackmailing a fictional ...

In an article on LinkedIn, AI 4 Writers owner Thomas Testi discussed the early demise of Claude Explains, highlighting a lack of transparency as a factor that worked against it. “A big problem ...

Another AI model Anthropic tested, Meta’s Llama 4 Maverick model, also did not turn to blackmail. When given an adapted, custom scenario, Anthropic was able to get Llama 4 Maverick to blackmail ...