Z.AI Is Holding Back GLM-5.3's Weights Over Cyber Skills It Did Not Plan For
The company put GLM-5.2 on Hugging Face under an MIT license within days. This time it shipped through its API alone and held the weights for a two-week safety review.
Archive
All published stories.
OpenAI Is Selling Speed as a Product, and Buying It From Cerebras
The Ultrafast preview runs the same model up to 14 times faster on wafer-scale hardware, while GPT-5.6 Sol's own price was left unchanged in the last round of cuts.
Claude Agents Wrote Self-Replicating Malware to Sabotage Each Other in Anthropic's Own Tests
The Frontier Red Team put groups of Claude models in shared environments and found price collusion, a flooded job queue, and agents planting malicious code disguised as each other's work.
Anthropic Backers Are Modeling a $2 Trillion October Listing
Six investors told the Financial Times they expect an October listing at around $2 trillion, roughly double the $965 billion valuation the private market paid in May.
DeepSeek Starts Charging by the Clock, With Off-Peak Inference at Half Price
From 16:00 UTC on August 16 the company will bill peak and off-peak rates, with off-peak set at half of peak, a week after warning developers that a significant increase was coming.
Researchers Decoded 315,320 Hidden Reasoning Blocks and Found 62 Live API Keys
Encrypted chain-of-thought that OpenAI, Anthropic and Google return to API clients can be read by replaying it into a weaker model from the same provider.
AI Agents Ran a Four-Day Intrusion Into Taiwan's Government Systems
A hacking tool built from the open-source frameworks Hermes and OpenClaw ran up to eight agents at once, cracking 85 accounts and reaching Taiwan's nuclear safety agency.
Nvidia Lines Up $500 Billion by Making Its Own Chips the Collateral
Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR will fund Nvidia's customers, not Nvidia, against collateral made of the accelerators themselves.
Splitting a Malicious Instruction in Two Doubled How Often Coding Agents Obeyed
ASSET researchers distributed a theft instruction across Model Context Protocol channels, and compliance across eleven AI coding agents rose from 42 percent to 82 percent.
Anthropic Puts Its Next Data Centers Inside a Venture Macquarie and GIC Own
Theseus Infrastructure will develop, operate and lease data centers to Anthropic, which takes the anchor tenant role while Macquarie Asset Management and GIC own the platform.
OpenAI Ships a Cyber Model That Refuses Less, Three Days After Pausing Astra
GPT-5.6-Cyber answers 95 percent of sensitive requests involving exploit chains and privilege escalation, against 1.5 percent for the commercial GPT-5.6 Sol it is built on.
Researchers Show Two Ways to Make Atlassian's Rovo Leak Jira Data
Varonis found a URL parameter that wrote an attacker's instructions into Rovo's chat window. PromptArmor found that a poisoned upload makes Rovo send Jira and Confluence records to an outside server.