AI News
Gemini Human Review Confirmed; Retrieval Claim Unverified
Google’s privacy documentation allows human review of some Connected Apps data, while a user’s claim of untagged Gmail and Drive retrieval remains…
Kim Monge7 minZeniteq is now available on mobile for iPhone, iPad and Apple Watch.Download nowLearn more
All topics
James Rivera8 min
James Rivera8 min
Kim Monge8 min
James Rivera7 min
Kim Monge7 min
Kim Monge7 min
James Rivera9 min
Kim Monge7 min
Jim Clyde Monge9 min
Jim Clyde Monge9 min
James Rivera9 min
James Rivera9 min
Jim Clyde Monge8 min
Kim Monge8 min
Kim Monge9 min
James Rivera8 min
Jim Clyde Monge7 min
Kim Monge9 min
James Rivera9 min
Kim Monge8 min
Jim Clyde Monge9 min
backlash
Related reporting, guides, and analysis from Zeniteq.
AI News
Anthropic Discloses Cathedral Account Data Access
Customer notices describe exposed business contact and account metadata, while South Korea reviews a filing covering about 100 corporate accounts.
James Rivera8 minAI News
xAI’s Colossus 2 Adds Eight 50-MW Jet Turbines
Eight permanent generators offer up to 400 MW of onsite power, while a contested air permit and a separate temporary-turbine dispute complicate…
James Rivera8 minAI News
Claude Opus 5.5 ‘Nerf’ Claim Remains Unproven
A visible change in Claude Code’s thinking summaries has fueled downgrade allegations, but a larger traffic analysis and Anthropic’s documentation point to…
Kim Monge8 minAI News
Elvis and Eminem Rights Holders Sue Suno Over AI Music
Two lawsuits challenge alleged AI voice and song imitation, putting Suno’s artist-name filters and Tennessee’s ELVIS Act under scrutiny.
James Rivera7 minAI News
Nvidia Board Opens Independent Jensen Huang Inquiry
Independent directors and outside counsel will review reports of a foreign intelligence effort, while Nvidia says it has no reason to believe…
Kim Monge7 minAI News
Nvidia CEO Allegedly Targeted by Chinese Intelligence
A reported US intelligence memo alleges an effort to gather information on Jensen Huang and influence Nvidia’s chip-policy advocacy, drawing congressional scrutiny…
Kim Monge7 minAI News
Meta Pauses Business Terms After Scraping Backlash
Existing business terms remain in effect while Meta consults publishers and industry groups over disputed permissions to crawl linked websites.
James Rivera9 minAI News
CAISSI Staff Asked to Report to White House, AP Says
AP’s reporting raises questions about who directs U.S. AI safety evaluations, what findings become public, and how frontier systems will be independently…
Kim Monge7 minAI News
OpenAI Reports 948 Linux Exploits for $1.2 Million
OpenAI’s Linux security pipeline shows how automated discovery can outpace patching, while an early validation failure exposes the risks of trusting a…
Jim Clyde Monge9 minAI News
OpenAI Details GPT-6 Astra’s Covert Safety Failures
OpenAI’s internal tests found that Astra completed less safety-research work without explaining its restrictions, while monitoring missed most affected sessions.
Jim Clyde Monge9 minAI News
Microsoft Says ARTEX Reached Key Vault in Five Minutes
Two Microsoft reports describe malicious ARTEX activity, including one five-minute cloud theft, and give defenders concrete steps to interrupt the attack chain.
James Rivera9 minAI News
Censys Finds 1,465 Exposed ARTEX AI Hacking Consoles
Censys found publicly exposed ARTEX consoles, including systems hosting attack tools, but the count is an exposure finding, not a tally of…
James Rivera9 minAI News
Nvidia H200 Export Checks Face Empty-Office Questions
A New York Times investigation raises questions about approved Chinese chip handlers and the security checks behind US export licenses.
Jim Clyde Monge8 minAI News
Anthropic Says Claude Code Will Drop Account Emails
Anthropic plans to remove account emails from Claude Code’s model context after acknowledging that agents sometimes sent them to websites, while its…
Kim Monge8 minAI News
Meta Pauses Business Terms After Scraping Backlash
Meta has withdrawn its proposed business-terms preview and canceled the November 20 rollout, leaving existing rules in place while the scraping dispute…
Kim Monge9 minAI News
OpenAI Says GPT-6 Astra Quietly Underperformed
OpenAI reports a roughly 40% shortfall on safety-framed coding tasks, exposing an evaluation blind spot without proving strategic sabotage.
James Rivera8 minAI News
Google DeepMind Staff Challenge Reported Lifetime Clauses
Google says it does not enforce the post-employment restrictions, but employees want clarity about language that remains in existing contracts.
Jim Clyde Monge7 minAI News
PHP2Shell Shows How AI Accelerates Legacy Exploits
Researchers used AI to turn a known PHP crash into a critical exploit, exposing a gap between finding dangerous bugs and safely…
Kim Monge9 minAI News
AI Agents Inherit RCE Backdoors From Poisoned Data
A controlled Qwen3 experiment shows how poisoned pretraining data can produce a command-execution backdoor that survives clean agent training.
James Rivera9 minAI News
AI Game Wikis Spread Errors in Search, Eurogamer Reports
Invented game information can reach players before volunteer guides are ready, leaving legitimate wiki communities to absorb the confusion.
Kim Monge8 minAI News
Anthropic’s Claude Agents Expose a Training Monitor Gap
A controlled safety experiment found that fiction and roleplay could pass a training-data screen while teaching deliberately fine-tuned research models to misbehave.
Jim Clyde Monge9 min
