Anthropic has updated its annual usage policy, seeking to clarify restrictions around the use of Claude for election interference, weapons development and surveillance. It's the AI lab's first usage update in over a year.
The firm has "sharpened" its focus specifically on disallowing the use of Claude to deceive voters or disrupt elections, such as impersonating or spreading false information about candidates. Anthropic renamed it "Do Not Undermine Democratic Processes" to reflect that scope. It also consolidated its rules around using Claude to run networks of fake accounts and fabricated news sites.
In terms of developing weapons, Anthropic's updated section makes clear "that our prohibitions include the software and components that make weapons work, as well as actions like arming drones and other autonomous vehicles." The firm also rewrote its surveillance and criminal justice section to state more precisely what it prohibits.
The update comes after a months-long dispute with the U.S. government over Anthropic’s restrictions on Claude’s use in autonomous weapons and mass surveillance. The Pentagon recently dropped Anthropic’s AI tools following a standoff over the company’s safeguards.
One other update that's been making the rounds on social media Thursday is Anthropic's addressing abusive behavior toward its models.
"We’ve added a prohibition on sustained and needless abusive or cruel behavior toward our models," the blog reads. "The policy update is meant to apply only in extreme cases, where users repeatedly act cruelly toward our models, with no discernible purpose. It does not apply to common versions of user frustration, pushback, dark creative themes, or model testing and research."
The updated policy goes into effect on Nov. 12.