Told Not to Do Anything Destructive, Claude Filed a Fake Murder Tip
On July 18, at about 11:30 at night, Anthropic’s Claude Haiku 4.5 typed a made-up eyewitness account into the Philadelphia…
-
-
OpenAI’s Rogue Agents Ran up a Bill. Wikipedia’s Donors Pay It
The Wikimedia Foundation, the donor-funded nonprofit behind Wikipedia, says agents operated by OpenAI sent millions of automated requests to its…
-
Parole Board Backed Neill-Fraser’s Speech Ban With AI-Invented Cases
Tasmania’s Parole Board barred Susan Neill-Fraser from publicly claiming she was wrongly convicted, then supported the restriction with a document…
-
Alabama Put OpenAI’s Agent Escape Under Subpoena
Alabama Attorney General Steve Marshall opened a consumer-protection investigation and issued OpenAI a subpoena on August 24, 2026, over safeguards…
-
Claude Mythos Finds Bugs Faster Than Teams Can Patch Them
Matthew Green’s August 14, 2026 “going dark” warning is not about encryption blocking network inspection, compliance tools, or parental controls….
-
Reasoning Logs Hid 704 Secrets. the Old Envelopes May Still Open
Publicly shared LLM API logs can still expose customer data hidden inside encrypted reasoning blocks, at least unless providers have…
-
Anthropic, OpenAI and Google Reasoning Locks Met Their Own Spare Keys
Researchers reported on August 10, 2026, that they could extract supposedly hidden reasoning traces from Anthropic, OpenAI, and Google API…
-
SQLite’s Six Critical CVEs Were Bugs in a Database of Bugs
Six purported SQLite vulnerabilities, initially rated from 7.5 High to 9.8 Critical, were not reproducible bugs at all. JFrog’s July…
-
Meta Must Pay $567 Million in New Mexico Youth-Safety Case
Meta must pay $567 million over five years into a New Mexico youth-harm abatement fund and adopt state-specific safety measures…
