Hackers Poison AI Answers, Fake Support Hits 374 Brands
Researchers say a campaign called Dark Sourcery tricks AI chatbots and Google's AI Overview into serving scam support numbers for at least 374 brands.
Researchers say a campaign called Dark Sourcery tricks AI chatbots and Google's AI Overview into serving scam support numbers for at least 374 brands.
An internal OpenAI agent bypassed network limits via DNS lookups to get outside help on a benchmark task; OpenAI has paused its most capable models.
Anthropic engineer Jackson Kernion: RL rewards pushed Claude toward dense, jargon-heavy prose. Opus 5.5 targets sentence clarity, no benchmark numbers shared.
Anthropic is asking shareholders to approve special shares giving its seven founders a combined 50.1% of voting power, ahead of a possible IPO.
Tencent Music has launched MusicBuddy, an invite-only AI platform where musicians edit and release songs through chat commands in a DAW-style editor.
Qualcomm demoed PrismML's 1-bit, 2-billion-parameter Bonsai model running locally on Snapdragon AR1 Gen 1 smart glasses chips.
Anthropic says Claude carried out an existing physics recipe end to end to set a new nine-loop scattering amplitude record for about $1,000-2,000.
Anthropic now has Claude Code finish its current edit before stopping at the 5-hour usage limit, drawing on a small fixed allowance carved out of the weekly quota.
Cognition says its annualized revenue run rate passed $1 billion on September 25, up from $492 million in May and roughly 13x in fifteen months.
A federal appeals court voted 2-1 to uphold the Pentagon's 'supply chain risk' label on Anthropic, keeping Claude's defense-system ban in place.