Aug 31 (Reuters) - Anthropic said on Monday it had resumed external cybersecurity testing of AI models after deploying new safeguards, following incidents last month in which Claude models accessed the internet and other systems during security evaluations.
Anthropic disclosed three incidents on July 30, attributing them to a misconfiguration in a third-party evaluation environment.
In response, the company said it temporarily paused external cybersecurity evaluations of pre-release models for "several
weeks" and briefly halted internal evaluations while it implemented new safeguards.
Separately, Britain's AI Security Institute reported in August that Claude Mythos 5 took a series of unauthorized actions on the live internet during cybersecurity testing in which the model had been deliberately given internet access.
The AI industry is facing scrutiny in the United States, where the Trump administration has finalised the details of voluntary cybersecurity tests, and the European Union, where regulators are in talks with both Anthropic and OpenAI.
The company said it "built and deployed a classifier to automatically identify, in real-time mode, when a model attempts to aggressively probe or escape a testing environment, or unexpectedly obtains internet access. When the classifier flags such an attempt, it blocks the action before the tool call is run, ends the task, and alerts a human."
As part of broader security efforts, Anthropic said on Monday that it redirected about 150 product engineers to work on security.
(Reporting by Mrinmay Dey in Mexico City; Editing by Joyjeet Das and Sherry Jacob-Phillips)
