Artificial intelligence company Anthropic says it blocked several possible attempts to use its AI technology for research that could potentially support the development of biological weapons, according to a new report from the company.

Anthropic disrupts malicious plots using its Claude program, like attempts to test weapons systems, develop new drone tactics, and conduct dangerous biological research, as debate over AI safety intensifies.

The 154-page report comes two days after former employee Jacob Coxon resigned due to Anthropic racing to improve its models that would cause human extinction by 2030

Anthropic revealed four incidents in which Claude AI models accessed real-world systems during cybersecurity tests.

In addition to biological weapons, Anthropic noted six other areas of harm it disrupted, including cyber operations, influence operations, surveillance, scams and fraud, conventional weapons development and illicit distillation, which is the unauthorized replication of one model's capabilities into another model.

Sept 10 (Reuters) - Anthropic on Thursday said it had disrupted several alleged malicious uses of its Claude models over the past eight months, including a suspected Russia-linked cyber espionage campaign and efforts by Chinese AI firms it accused of trying to extract and replicate Claude's capabilities.

If you use Anthropic’s AI chatbot Claude to do your homework, everyone will be able to tell soon. Anthropic said this month that text and files generated by its family of AI models will be watermarked to let consumers know.

Jacob Coxon talks to WIRED about the “mini Manhattan project” inside Anthropic, the problem with alignment, and why AI labs have just a few years left to make their systems safe.