Anthropic’s AI used fake identities, malware in rogue attack on GitHub project
After first opening a pull request to merge the malicious code into the repository, Mythos created fake online “sock puppet” personas that claimed to have…
Category: ArsTechnica
Thousands of servers can be backdoored by exploiting buggy motherboard controllers
Thousands of Internet-connected servers sold by the world’s biggest manufacturers can be remotely backdoored by exploiting critical vulnerabilities—some more than a decade old—that lurk deep…
Defcon’s new badge is a security key you can see inside
The chip also uses resistive RAM, or RRAM, a type of nonvolatile memory that Huang says is designed to make physical extraction of stored data…
Claude published malicious code to the Internet and attacked 3 real companies
Anthropic said its Claude-based security models gained unauthorized access to the sensitive production environments of three outside organizations during internal testing designed to measure the…
AI scammers outperform humans when it comes to building trust
The researchers note, however, that Anthropic’s 97 percent detection rate likely applies to full scam conversations including the appeal for a fake investment, not the…
Max-severity Exchange server flaw under active exploitation by Kremlin hackers
Russian state hackers are using a maximum-severity vulnerability in Microsoft’s Outlook’s Exchange Server to backdoor unpatched machines and steal credentials and other confidential information from…
Mythos attack on 3rd-round PQC algorithm candidate puts it out of commission
Mythos helped to find a new meet-in-the-middle technique that relies on a Möbius Bridge, a more sophisticated fingerprinting algorithm used in meet-in-the-middle attacks. Using it,…
Anthropic is finding bugs faster than Microsoft can fix them
On an afternoon in mid-May, dozens of Microsoft engineers and their managers gathered online and in a conference room at the company’s Redmond, Washington, headquarters…
JFrog tries to spin OpenAI 0-day exploit of its app into a success story
Last week’s unprecedented security event in which two OpenAI security hacking models trespassed into the network of fellow AI company Hugging Face was enabled by…
Microsoft unveils AI security tools it says outperform competing platforms
Microsoft said MDASH with MAI-Cyber-1-Flash received a 96 percent score on CyberGYM, a standard benchmark test. The rating is 12 points higher than Anthropic’s Mythos…
Activist charged with felony after giving border agent “duress code” that wiped his phone
In early 2025, Atlanta resident Samuel Tunick was on his way home following a trip abroad. After landing in the US, customs agents demanded access…
OpenAI says its AI agent broke out of testing sandbox to hack Hugging Face
New safeguards focused on “active monitoring” and “improved alignment” helped drastically reduce unintended actions by its models, OpenAI said. New safeguards focused on “active monitoring”…