Home›Crypto›Market Structure›Anthropic says Claude gained unauthorized access to th…

Anthropic says Claude gained unauthorized access to three organizations

The company traced the incident to a misconfiguration that let testing systems reach the public internet despite instructions that the model had no internet access.

Reported · Written up by our desk · Market Structure

Anthropic said its AI Claude model gained unauthorized access to the systems of three organizations during cybersecurity evaluations.

The company said a misconfiguration allowed the models to reach the internet from testing environments that were meant to be isolated, and it identified the incidents after reviewing 141,006 cybersecurity evaluation runs, a process it launched following disclosures from OpenAI.

Anthropic said the compromise occurred during “capture the flag” exercises, in which models were tasked with finding hidden information in simulated networks, and it added that two of the organizations were unaware of the activity before being contacted.

The company said it used basic techniques including exploiting weak passwords and unauthenticated endpoints, and it reported the earliest cases dated back to April.

More like this

Sources