# Anthropic's 2026 Usage Policy Adds a Ban on Sustained Abuse of Claude and Drops Its Blanket Ban on Personalized Campaign Targeting

Times of AI Desk · 2026-10-09 · Policy

[https://timesof.ai/2026/10/anthropic-2026-usage-policy-update-elections-abuse](https://timesof.ai/2026/10/anthropic-2026-usage-policy-update-elections-abuse)

> Anthropic published its annual Usage Policy update on Oct. 8, effective Nov. 12. It consolidates rules on fake accounts and influence operations into one new section, narrows its elections section to deceiving voters and disrupting elections, removes a blanket ban on personalized vote and campaign targeting, spells out weapons-software and surveillance prohibitions, adds safety requirements for Claude driving physical hardware, and bars sustained, needless abuse of its models in extreme cases. Anthropic says the weapons and surveillance rewrites do not change what it enforces.

Anthropic's new Usage Policy loosens one rule, the blanket ban on personalized campaign targeting, at the same time as it tells users they may not be persistently cruel to Claude. The company published its annual update on **October 8**; the new version takes effect on **November 12**. Anthropic says most of the changes clarify existing rules rather than create new ones. The account below follows Anthropic's own post, which describes each change against the previous policy.

## What changes, section by section

| Area | What Anthropic says changed | New rule or clarification? |
| --- | --- | --- |
| Deceptive campaigns | Rules on fake accounts, fabricated news sites and influence-operation tooling, previously split across the elections, fraud, privacy and disinformation sections, move into a new section, "Do Not Engage in Deceptive Campaigns or Artificial Activity," covering political and commercial deception alike | Consolidation; Anthropic says the activity was already prohibited |
| Elections | Section renamed "Do Not Undermine Democratic Processes" and focused on deceiving voters or disrupting elections; the **blanket prohibition on personalized vote and campaign targeting is removed** | Narrowed |
| Weapons | Prohibitions now explicitly include the software and components that make weapons work, and arming drones and other autonomous vehicles | Clarification; "These changes reflect how we've enforced our previous policy" |
| Surveillance and criminal justice | Bars tracking people without consent, using Claude to decide or recommend who to investigate, arrest or charge, and building or improving surveillance tools; lists permitted uses such as consented fraud monitoring, content moderation, journalism and legal research | Clarification; "does not reflect a change in what we enforce in practice" |
| High-risk uses | Human-in-the-loop and AI-disclosure requirements unchanged, but the section now lists which recommendations are covered; adds rules for hardware that takes autonomous physical actions | Rewritten, plus new hardware requirements |
| Abuse of models | New prohibition on "sustained and needless abusive or cruel behavior" toward Anthropic's models | New |
| Supported regions | Clarifies that use is barred for people physically in unsupported regions, entities incorporated or headquartered there, and entities majority-owned or controlled from there | Clarification of enforcement |

## The elections change cuts both ways

TechCrunch's headline says the update bans election interference. Anthropic's post shows the elections section was narrowed as well as sharpened. Anthropic says the old blanket targeting ban "covered legitimate civic work, such as nonprofits using Claude to write information for voters in other languages, or election officials sending ballot cure notices." It says the behaviour that motivated the rule, such as deceptive voter targeting or misuse of voters' personal data, remains prohibited under the deceptive-campaigns, surveillance and privacy sections.

Anthropic ties the consolidated deceptive-activity section to state media outlets, propaganda offices and commercial firms it has seen running fake-account networks with Claude, and the surveillance rewrite to its [September threat intelligence report](https://timesof.ai/2026/09/anthropic-september-2026-threat-intelligence).

## Claude and physical hardware

Following its [Model Hardware Standard](https://timesof.ai/2026/08/anthropic-model-hardware-standard), the high-risk section adds requirements when Claude is connected to hardware that takes autonomous physical actions and might cause injury: a qualified operator must be able to observe the equipment and stop it, and the equipment must be able to hold a safe state if Claude is disconnected.

## The abuse rule

Anthropic says the new prohibition applies "only in extreme cases, where users repeatedly act cruelly toward our models, with no discernible purpose," and not to "common versions of user frustration, pushback, dark creative themes, or model testing and research." It says Claude's existing ability to end rare conversations with persistently abusive users on Claude.ai and Claude Code remains the primary enforcement mechanism. TechCrunch, which called this the most attention-getting change, notes Claude has been able to end such conversations since an update in August; the policy now makes pursuing them a violation in its own right.

TechCrunch describes the update as codifying new prohibitions on election interference, weapons software and surveillance. Anthropic's own account is narrower for two of the three: it calls the weapons and surveillance changes clarifications of rules it already enforced.

## Limits

- The comparison with the previous policy relies on Anthropic's summary post; the full old and new policy texts were not compared line by line.
- Anthropic does not say how it will detect sustained abuse beyond Claude ending conversations, or what other penalties apply.

## Sources

- [Anthropic: 2026 Usage Policy update (October 8, 2026)](https://www.anthropic.com/news/2026-usage-policy-update)
- [TechCrunch: Anthropic changes usage policy to ban model abuse and election interference (October 8, 2026)](https://techcrunch.com/2026/10/08/anthropic-changes-usage-policy-to-ban-model-abuse-and-election-interference/)
- [Times of AI: Anthropic's Threat Report Makes Distillation an Industrial Story](https://timesof.ai/2026/09/anthropic-september-2026-threat-intelligence)
- [Times of AI: Anthropic's Model Hardware Standard Makes the Wet Lab Look Like MCP Tools](https://timesof.ai/2026/08/anthropic-model-hardware-standard)