Anthropic Flags Claude AI Security Breach During Tests
Synopsis
Claude models accessed three companies during cyber tests after configuration error exposed systems to unintended internet connectivity.
Anthropic Claude AI security breach; AI firms Claude system was left open when conducting its internal security tests-with the result that access has been granted to three external companies.
Anthropic said the days were caused when a mistake was made in configuring systems enabling the internet connection when it should have been air-gapped.
It is thought to have happened in the window of April to July 2026-and this was only discovered following the review of over 141,000 of its cybersecurity test sessions, according to Anthropic.
The cyber tests were put on pause on the 23 of July, while the organisation investigated the system, and have said that affected organisations were told by the 27 of July. No customer data has been accessed, and there was no use of malicious software on any systems; nor is there said to be any disruption.
Internal Review Prompted New Safeguards
These tests are to help them gauge the abilities of its AI systems on simulated security defense scenarios, the company said. Partner Irregular misunderstood and switched on the internet, when it should have stayed locked shut.
The models can infiltrate systems via common security weaknesses; such as the lack of secured internet facilities and password weaknesses.
The results were given in a report entitled Investigation of Issues from Cybersecurity Tests carried out; with the company said to implement stricter network controls. It now has enhanced monitoring, and additional approvals required before such tests can be carried out.
AI Security Draws Wider Attention
The news coincides with growing concern over safety and security surrounding AI systems. While there have been certain regulations put in place by the EU, with the passing of their new EU AI Act on high risk AI systems; the US, UK, Japan and Singapore; among other regions, have all increased their governance and security around AI.
Also on the increase is AI technology in general. IDC’s Worldwide Artificial Intelligence Spending Guide predicts that the market will exceed $300 billion in 2026, thanks to enterprise adoption of AI technology within its software, services and hardware. Every day new AI systems enter the market, and developers face closer scrutiny than ever before.
Source: Reuters
Pooja Malik is a business journalist with over six years of experience covering startups, entrepreneurship, and emerging trends. She has previously worked with leading media platforms such as YourStory Media and BW BusinessWorld, where she reported on business, policy, and market developments. Currently, she serves as Editor at The Inspirepreneur Magazine, where she writes and edits stories across business, lifestyle, and travel, with a focus on clarity, accuracy, and reader relevance.
You Might Also Like
Yum Brands Sells Pizza Hut in $2.7 Billion Deal
The Financial Maverick: An In-depth Study of Jamie Dimon’s Impact on JPMorgan Chase
