RC RANDOM CHAOS

The Wire

Curated cybersecurity and tech news — AI-summarized, source attributed.

aiopen-source

Academic Lab Bets on Local AI: Frontier Models on a Single 24GB GPU

Tim Dettmers argues that the future of AI research belongs to resource-constrained university labs, not GPU-rich frontier companies, and his group is making the

via Hacker News ·
aicybersecurity

AI cracks a 1941 Enigma message that stumped cryptanalysts for 20 years

A German Army Enigma message from 10 July 1941 — logged as Nr. 172, callsign MVUEH — that had resisted every attempt at solution since 2005 has finally been bro

via Hacker News ·
cybersecurityvulnerability

AMD's RDRAND may never return a true zero, assembly hobbyist claims

A user on the flat assembler forum reports that on the AMD chips he tested, the RDRAND hardware random-number instruction never produces a full-width zero at th

via Hacker News ·
aicybersecurity

Anthropic ships Opus 5.5: Fable 5.1-class work at 40% lower cost

Anthropic has launched Claude Opus 5.5, the first model in its 5.5 family, positioning it as a new frontier leader that matches the more expensive Fable 5.1 on

via Hacker News ·
privacyai

Apple Strips Out the Off Switch for On-Device AI in Latest macOS

A blogger recounts how Apple has steadily eroded user control over its AI features. In early 2025, macOS 15.3 quietly turned on a feature that transmitted perso

via Hacker News ·
tech-cultureopen-source

Apple's Cancelled Copland OS Runs in the Browser via WebAssembly

Copland, the microkernel-based operating system Apple built in the 1990s to succeed System 7, was scrapped before shipping and has been almost impossible to run

via Hacker News ·
aitech-culture

Claude Opus 5.5 Tops Intelligence Index at 58 — Fast, Verbose, and Costly

Anthropic's Claude Opus 5.5, released September 22, 2026, lands near the top of Artificial Analysis's Intelligence Index with a score of 58, more than double th

via Hacker News ·
aicloud

Claude Recovers from 80-Minute Outage Spanning Opus 5, Fable, and Mythos Models

Anthropic experienced elevated error rates across several Claude models for roughly 80 minutes on September 22, with impact running from 5:50pm to 7:10pm PT (00

via Hacker News ·
clouddevops

Cloudflare makes Python Workers generally available, with native bindings and DB drivers

Cloudflare has promoted Python Workers to general availability, two years after first letting developers run Python inside its edge Workers runtime. Python is n

via Hacker News ·
policytech-culture

Disney+ terms let ads run in 'ad-free' tiers — now before and after movies

A January 2025 revision to the Disney+ Subscriber Agreement (Section 2(k)) carved out exceptions that let advertising appear even in tiers marketed as 'no ads'

via Hacker News ·
aipolicy

Elite mathematicians form independent group to police AI's math claims

A group of prominent mathematicians — including Fields Medalists Timothy Gowers and Martin Hairer, physicist Edward Witten, and Ravi Vakil — has launched the Ad

via Hacker News ·

Fable 5 user says the model got dumber after Anthropic bundled it into subscriptions

A widely shared post claims Anthropic's flagship Fable 5 model degraded noticeably once the company made it permanently available inside subscription plans. The

via Hacker News ·