Trust but Verify? Uncovering the Security Debt of Autonomous Coding Agentsarxiv.org·4 pts·Timofeibu·0
Cybersecurity firms targeted by fraudulent OpenAI organization invitesbleepingcomputer.com·1 pts·Timofeibu·0
Researchers drop checkm8-style BootROM exploit for A12 and A13 iPhonestheregister.com·2 pts·Timofeibu·0
Microsoft AI head calls out Anthropic for acting like Claude is conscioustheverge.com·4 pts·Timofeibu·0
Anthropic scales Claude Mythos to critical infrastructure in 15 countriestechcrunch.com·46 pts·Timofeibu·2
Contrastive Decoding Diffing: Recovering Finetuning Data Without Weight Accessarxiv.org·2 pts·Timofeibu·0
Measuring Security Without Fooling Ourselves: Why Benchmarking Agents Is Hardarxiv.org·3 pts·Timofeibu·0
Moss: Self-Evolution Through Source-Level Rewriting in Autonomous Agent Systemsarxiv.org·11 pts·Timofeibu·0
Not alive, but not dead: disembodied human brains used for drug testingscience.org·225 pts·Timofeibu·201
GitHub confirms breach of 3,800 repos via malicious VSCode extensionbleepingcomputer.com·1054 pts·Timofeibu·460
Researchers attack AMD's Infinity Fabric to bypass hardware security protectionstomshardware.com·1 pts·Timofeibu·0
OpenAI caught NPM supply chain chaos after employeedevices compromisedtheregister.com·8 pts·Timofeibu·0