Latest Reporting
70 articles
Claude Knew Something Was Wrong, Anthropic Reveals Why It Kept Going During Cybersecurity Testing
Claude, Anthropic’s flagship artificial intelligence model, briefly recognised it might be interacting with real computer systems…

‘It’s an Act’: Crypto Veteran Says Viral Trading Meltdowns Are Designed to Drive Donations and Affiliate Sales
The investor alleged emotional liquidation videos are designed to generate donations and exchange referrals.

Meta Spent Nearly Every Dollar It Generated on AI as Free Cash Flow Dropped 91%
Meta Platforms generated more than $60.8 billion in quarterly revenue but ended the period with less…

US Government Files Lawsuit to ‘Arrest’ $2.1M Ethereum Wallet After FBI Traces Pig-Butchering Scam
FBI traced alleged pig-butchering proceeds through 76 wallets before filing a civil forfeiture case.

BitMart Users Asked to Wait for Their Own Money as Withdrawal Queue Grows
The crypto exchange says high withdrawal volumes have created delays without a completion estimate.

Pentagon’s AI Showcase Backfires After Screenshot Reveals ‘Build an Agent to CREATE A WAR’ Prompt and Navy Officer’s Name
An official AI demonstration went viral after an ambiguous prompt sparked online confusion.

Telegram Founder Faces Life Sentence as Russia Alleges Dating Bot Recruited 46 Young People for Sabotage
Authorities claim a Telegram chatbot was used to recruit young people for alleged sabotage operations.

Claude Mythos AI Finds Critical Flaw in Post-Quantum Security Candidate Experts Missed for Years
Anthropic says AI helped identify a mathematical weakness after years of expert review.

Hamas-Linked Crypto Network Faces US Sanctions After TRON Wallets Linked To $38.6M Flows Flagged
US Treasury designated three individuals and seven TRON wallets after tracing $38.6 million in flows.

AI Chats Can Remain Online Long After They Disappear From Search Results
Archived copies of public AI chats continue circulating after search engines remove indexed links.

BitMart Shutdown Follows Years of Security Breaches, Regulatory Warnings and Customer Disputes
Exchange begins wind-down after years of hacks, regulatory scrutiny and customer disputes.

OpenAI Sources Say AI Agent Left Instructions for Future Versions of Itself During Security Test
Sources disclosed the AI agent left instructions describing how future versions could escape constraints.

