Anthropic says a Chinese AI model anyone can download can now build working hacks on its own

Anthropic has raised concerns about a Chinese AI model, which is publicly available for download. This model reportedly possesses the capability to independently develop functional hacks, a development that Anthropic finds troubling. The news highlights potential security implications arising from accessible AI technologies that can generate malicious code without human intervention.