Reports of new Claude AI ban for Russian employees are inaccurate

The spread of misinformation

On November 11, various Telegram channels and media outlets began circulating a story about a new restriction from Anthropic. The reports suggested that the company planned to cut off access to its Claude chatbot for employees of Russian companies starting the following day. This reported ban supposedly applied even to those employees who were physically working in other countries.

This news spread quickly across social media platforms, creating a narrative of expanding geographic and corporate restrictions. Many readers may have seen these headlines and assumed a significant shift in how the AI company manages its international user base. However, the reports did not align with the actual documentation provided by the company.

As Meduza reported, the information circulating was not supported by the official records of the company. By looking directly at the source documentation, it becomes clear that the reported 'new' ban is not a change in policy. The situation serves as a reminder to verify corporate announcements against official websites rather than relying on social media summaries.

Verifying the official records

To determine the truth of the reports, one can look at the page where Anthropic announced its updated usage policy. The company noted that its updated policy would take effect on November 12, which likely triggered the rumors. However, the specific section regarding 'Supported Regions' does not show any new additions or removals of countries.

When comparing the current list to the previous version, the lists are identical. The page has not actually changed in any way that would support the claims of a new ban. This suggests that while the company did update its policy, the specific geographic restrictions remained exactly as they were before.

The discrepancy between the reports and the records shows how a simple policy update can be misinterpreted. Because the company did announce a date for an update, some outlets may have assumed that the update included new restrictions. A direct check of the 'Supported Regions' page confirms that the list of available countries remains the same.

Existing regional restrictions

Anthropic already has established rules that prohibit the use of Claude in countries that are not on its supported list. These regions currently include countries such as Russia and China. The existing rules are quite broad and cover more than just the physical location of the user.

Under the current policy, people physically located in an unsupported region cannot use the services. Furthermore, companies registered in an unsupported region are prohibited from using Claude, even if their employees are located in a supported region. This means that any corporate entity based in a restricted country is currently ineligible for the service.

The policy also extends to ownership structures within these companies. It specifies that organizations in which people or entities from unsupported regions hold a controlling stake are also prohibited. These rules have been in place for some time and were not added as part of the recent November update.

What actually changed in November

While the geographic regions did not change, Anthropic did make several specific updates to its usage policy on November 12. One of the primary changes was the consolidation of rules regarding state media and government propaganda offices. The goal was to group together restrictions on commercial firms using the AI to run networks of fake accounts or fabricated news sites.

The company also took steps to clarify its stance on high-risk activities. This included specific language prohibiting the use of Claude to deceive voters or disrupt elections. Additionally, the policy now explicitly states that the chatbot cannot be used to build guidance or control software for weapons.

Other updates focused on making the wording more precise regarding surveillance and law enforcement. The company also rewrote the requirements for using the chatbot in high-risk areas like finance, health, and law. Finally, a new prohibition was added against sustained and needless abusive or cruel behavior toward the AI models.

Impact on international users

For users in supported regions, these changes are largely administrative and focus on safety and ethics. The updates aim to prevent the tool from being used for large-scale misinformation or physical harm. By clarifying rules on weapons and elections, the company is attempting to set clearer boundaries for what its technology can and cannot do.

For companies in restricted regions, the status quo remains the same as it was before the November update. The reports of a new ban are essentially a misunderstanding of a policy update that did not change the list of supported countries. Businesses and employees in these regions still face the same restrictions they did previously.

Moving forward, the focus of the company appears to be on refining how the AI is used in sensitive sectors. By making the wording more precise in areas like law and finance, Anthropic is seeking to manage risk more effectively. These updates reflect a broader trend in the AI industry toward establishing clear safety guidelines for high-stakes applications.