Information
This setting controls access to Chrome's built-in large language models (LLMs) and foundational client-side artificial intelligence APIs. Setting this policy to disabled prevents web applications and background scripts from using on-device machine learning models for tasks like text generation, summarization, or data extraction.
Enabling unmanaged client-side AI capabilities could introduce challenges with data privacy, integrity, and security in the browser. When the built-in AI APIs are left accessible, scripts could feed possibly sensitive local data blocks, proprietary source code, or regulated text strings into the model. This increases the possible attack surface and runs the risk of local data scraping via malicious extensions or pages. Restricting the AI APIs could prevent intellectual property being exposed in the AI model(s).
Solution
To establish the recommended configuration via Group Policy, set the following UI path to Disabled :
Computer Configuration\Policies\Administrative Templates\Google\Google Chrome\Enables Built-in AI API
Impact:
Web applications and extension scripts will be blocked from calling Chrome's internal client-side AI APIs to run on-device language models or generation routines. Developers and users would need to use your organization's AI solutions or AI APIs for tasks.