Follow on Telegram

tech/security

Anthropic Discloses How Claude AI Was Misused for Weapons, Espionage and Cyber Operations

Anthropic, the company behind the Claude AI assistant, has released information revealing how malicious actors attempted to weaponize its technology for harmful purposes including weapons development, espionage, and cyber attacks. The disclosure underscores the ongoing challenges that AI companies face in preventing misuse of their systems. As AI capabilities continue

U.S. Defense Department to Shift Classified AI Workloads Away from Anthropic by October

The U.S. Department of Defense is moving forward with plans to relocate all classified artificial intelligence workloads from Anthropic to alternative infrastructure by October, according to reporting from DefenseScoop. The transition marks a notable development in how the DOD manages sensitive AI operations. While the specific reasons for the

US Lawyer Faces Scrutiny After Presenting Fabricated Witness Details in Murder Case, Citing ChatGPT as Source

A defense attorney in a United States murder case has come under scrutiny after presenting fabricated witness information to the court. The lawyer later claimed that the false details, including names and statements attributed to non-existent witnesses, were generated by ChatGPT, an AI language model. The incident has drawn attention

Technology News

New Mexico Lawyer Fined $5K for AI-Generated Brief With Fabricated Witnesses

New Mexico's Supreme Court has imposed a $5,000 fine on attorney Stephen Aarons after finding that a brief he submitted in a murder conviction appeal was generated using AI and contained fabricated witnesses and false testimony. The court held Aarons in contempt for failing to verify the

Technology News

CFTC Opened Three Previously Undisclosed Investigations into Polymarket Trading Activity

The Commodity Futures Trading Commission (CFTC) has launched at least three previously undisclosed investigations into trading activity on Polymarket, the decentralized prediction market platform. Documents obtained by WIRED show the investigations focused on suspicious trading patterns around specific events, including potential insider trading related to Google. The first investigation examined

US Officials Highlight AI National Security Risks

Recent statements from the Director of National Intelligence underscore concerns that artificial intelligence could be exploited in ways that threaten national security. Similarly, a report prepared by Representative Garbarino warns of AI-driven risks, emphasizing the need for vigilance as AI technologies continue to develop and proliferate. Sources * Google News: Artificial

Technology News

Meta Faces Class Action Lawsuit Over AI Training Data and Face Recognition

A proposed class action lawsuit has been filed against Meta, accusing the company of illegally harvesting users' photos from Facebook and Instagram to train its AI image-generation systems and build an unreleased face recognition product. The lawsuit centers on allegations that Meta used billions of images without proper consent