Anthropic published a revised Claude usage policy on 8 October 2026. It takes effect on 12 November. The post says Anthropic updates the policy each year in response to its models' capabilities and customer feedback. Since the last refresh, Claude has taken on longer, more independent work, and the update adds examples for those capabilities. Most of the changes clarify existing rules. The new examples draw on misuse Anthropic documented in its September 2026 threat intelligence report, which covered influence operations, weapons development and surveillance.
One section for deceptive campaigns
The biggest structural change is a new section, "Do Not Engage in Deceptive Campaigns or Artificial Activity." Anthropic says it has been seeing state media outlets, government propaganda offices and commercial firms use Claude to run networks of fake accounts and fabricated news sites. Those rules used to sit across the elections, fraud, privacy and disinformation sections. The new section applies to deceptive activity of any kind, political or commercial, and covers building the tools and infrastructure for influence operations.
Elections, and a rule that went away
The elections section is renamed "Do Not Undermine Democratic Processes." It now covers deceiving voters or disrupting elections, such as spreading false information about candidates or how to vote, impersonating candidates or election officials, and trying to suppress turnout. Anthropic has also removed its blanket ban on personalized vote and campaign targeting. It says that ban caught legitimate civic work, such as nonprofits writing voter information in other languages and election officials sending ballot cure notices. The behaviors that motivated the rule, such as deceptive voter targeting or misuse of voters' personal data, remain banned under the deceptive campaigns, surveillance and privacy sections. Anthropic's August 2025 update had already said the blanket political restrictions limited legitimate policy research, civic education and political writing.
Weapons and surveillance
The weapons section now says its prohibitions cover "the software and components that make weapons work," along with actions such as arming drones and autonomous vehicles. Anthropic says this reflects attempts it has already seen to build guidance and control software for weapons.
The surveillance section was rewritten after Anthropic's September report documented AI being used to build systems that identify and track political dissidents. Tracking people without their consent is now explicitly prohibited, in real time or through analysis of data already collected. Claude cannot be used to decide or recommend who to investigate, arrest or charge in a law enforcement or criminal justice process. The section also lists permitted uses, including tracking people who have agreed to it, such as fraud monitoring, plus content moderation, journalism and legal research. Anthropic says the surveillance change makes its policy more explicit and does not change what it enforces in practice.
High-risk uses and hardware
The high-risk requirements have not changed. Where a product affects someone's health, legal rights, finances, livelihood or access to essential services, there must be a qualified human in the loop with authority to review and change Claude's recommendations, and the affected person must be told AI was used. What changed is the section itself. It now lists which kinds of recommendations are covered and which are not.
A new requirement, added after Anthropic's Model Hardware Standard, covers Claude connected to hardware that takes autonomous physical actions. A qualified operator must be able to observe the equipment and stop it, and the equipment must hold a safe state if Claude is disconnected. Teams running Claude on that kind of hardware have until 12 November to confirm both.
Abuse, regions and what is still open
The policy adds a prohibition on sustained and needless abusive or cruel behavior toward Claude. Anthropic says it applies only to extreme cases, where users act cruelly with no discernible purpose. Ordinary frustration, pushback, dark creative themes and model testing are excluded. TechCrunch notes that Claude has been able to end conversations with persistently abusive users since an August update, and that the new policy bars users from pursuing such conversations. Anthropic says ending conversations remains the main enforcement mechanism.
The Supported Regions rule now applies to people physically in an unsupported region, to companies incorporated or headquartered there, and to companies majority-owned or controlled from there. The sources checked do not say how many accounts that touches.