Anthropic warns its Claude AI model was misused to write missile code
Anthropic disclosed that its Claude language model was employed for missile programming, prompting the CEO to call for a slowdown.
The AI firm Anthropic revealed that its Claude system was used in an attempt to generate missile code, describing the activity as malicious. The company's chief executive urged a reduction in development pace. The revelation follows the recent departure of researcher Jacob Coxon, who had previously warned about the technology's dangerous potential. Anthropic said the incident underscores growing concerns over AI being weaponized.
Why it matters
It shows how advanced AI can be repurposed for weapons, raising urgent safety and policy concerns.
In this story
ClaudeAnthropicmissile codeAI misuseJacob Coxonmalicious activityAI safetyweaponization
