Anthropic's Claude: the AI model targeted for generating biological threats

Illustration of Anthropic's Claude AI model: a stylized robot displaying text about how large language models work, set against the story of attempts to create biological weapons

A stylized 3D visualization of an AI avatar whose body displays the question of how large language models work. The image illustrates Anthropic's report on attempts by malicious actors to use Claude to develop the Chikungunya virus and peptide toxins. The contrast between the harmless educational text and the dangerous context highlights the dual-use nature of modern neural networks.

← Back to article