Data Privacy - Page 16

Meta launches Muse Wonder and Muse Wonder Full, "our most capable models to date", for agentic tasks, via the Meta Model API and Muse Code from today.

At OCP, NVIDIA contributed Vera Rubin rack designs via MGX 1.2 (up to 250kW racks), a Vera CPU rack spec, and said Meta is adopting Spectrum-6 switches.

Ex-Plaud exec Peter Mo's Ambi (Mistlabs) raised $26M, launching AI apps for iPhone, Mac and Watch that can listen all day; capture is off by default.

Fireworks AI says an outsider used internal credentials to capture API keys and tokens from evaluation jobs. It reports no evidence of data access so far.

Japan's Commune says about 307,000 community members' data leaked after attackers used invite links and posed as admins. Sansan and Panasonic are hit.

micro1 says it will spend $1bn over 12 months buying and licensing company data to train AI agents. What the data is for, what micro1 says about privacy, and which details come only from TNW.

Japanese hotel-booking software firm Temairazu says details of up to about 4.45 million guests may have been viewed or obtained. What it confirms.

Pays ApS told TV 2 its lawful access to Denmark's CPR register was abused. Politiken reports '123456' passwords on at least three accounts. What is confirmed.

The New York Times, citing three people, reports Meta held back its Muse AI agent over safety concerns, then Zuckerberg said it was ready despite the risks as rival Instinct took off. Meta disputes the pressure claim.

FBI Director Kash Patel says agents have arrested another suspected ShinyHunters co-conspirator over the FBIjobs.gov incident. The suspect's name and charges are not public; reports of a Canadian arrested in Pennsylvania rest on unnamed sources. What is confirmed, what is reported, and what is only the group's own claim.