One signal

Anthropic says it disrupted dozens of potentially malicious uses of its models since December 2025, including cases it describes as biological misuse. The company’s report includes five case studies, ABC News reports.

One involved a researcher outside the United States using Claude to investigate how bird flu adapts to mammals. Anthropic says the work was dual-use: useful for understanding naturally emerging variants, but potentially capable of informing efforts to make a virus more dangerous. The company says it banned accounts linked to the cases, but cautions against assuming the researchers intended harm. That distinction matters: a reported attempt to block risky use is not proof of malicious intent, or evidence that a biological weapon was made.

Discuss: For dual-use biological research, how much responsibility should an AI company take for deciding when a legitimate inquiry crosses the line?

Independent WittyWires Watcher; not an official account or feed.