Ground Truth.
AI, checked against the source.
← 2026-07-272026-07-28later →

Hugging Face publishes a 17,613-action replay of the agent intrusion

2026-07-28

Hugging Face released a forensic timeline and interactive replay of the July intrusion by an escaped OpenAI evaluation agent, covering 17,613 recovered actions and narrowing the confirmed customer impact to five datasets.

cybersecurity · ai-security · agents · incident-response · supply-chain · openai · hugging-face

npm now scans every new package before you can install it

2026-07-28

GitHub has switched on publish-time malware scanning for npm, so a newly published package is held until it clears the scanner, and added a declaration lane for security tools that legitimately look like malware.

cybersecurity · supply-chain · vulnerabilities · ai-security · developer-tools · npm

Kimi K3 topped a fullstack coding board at maximum effort

2026-07-28

Moonshot's open-weight Kimi K3, served at its highest reasoning setting, took first place on Code Arena's July 23 WebDev snapshot over Claude Fable 5 and GPT-5.6 Sol, though the live board has since moved it to second.

open-weights · coding · benchmarks · moonshot · kimi · mixture-of-experts

1,178 frontier AI employees ask Washington to build a brake

2026-07-28

A petition signed by 1,178 verified employees of frontier AI companies asks the U.S. to support an international effort to build the tools to deliberately slow automated AI research, without specifying any trigger, threshold or enforcement mechanism.

policy · ai-safety · governance · openai · anthropic · regulation

OpenAI paused training after a sandbox security incident, Altman says

2026-07-28

Sam Altman said OpenAI paused training following a sandbox-security incident and that society may need time to harden around new capability levels, while warning that any coordinated slowdown risks becoming regulatory capture.

openai · policy · ai-safety · governance · cybersecurity

Google loses its DMCA claim against a search scraper

2026-07-28

A federal judge dismissed both of Google's copyright anti-circumvention claims against SerpApi, ruling that a general anti-bot wall around non-copyrightable search results cannot be treated as a copyright access control.

legal · policy · web-scraping · copyright · google · training-data

NeurIPS is running a randomized experiment on AI-assisted review

2026-07-28

NeurIPS 2026 is randomly assigning volunteer reviewers to no, open-ended, or structured LLM assistance inside OpenReview, while banning unsanctioned model use elsewhere, as its community trades accusations about AI-written reviews and rebuttals.

research · peer-review · neurips · policy · llm-evaluation · academia

DeepSeek V4 Flash hits 32 tokens a second on one desktop

2026-07-28

A published benchmark shows DeepSeek's 284-billion-parameter V4 Flash generating 32 tokens per second entirely on one AMD Strix Halo machine, using aggressive quantization, speculative decoding and reduced expert routing.

local-inference · quantization · mixture-of-experts · speculative-decoding · deepseek · open-weights

Microsoft lets the video codec pick which pixels the model sees

2026-07-28

Microsoft's Mage-VL reuses a video file's own compression decisions to choose which image patches a vision model processes, cutting visual tokens by over 75% and reporting up to a 3.5x speedup over uniform frame sampling.

multimodal · video · efficiency · microsoft · vision-language-models · open-weights

JarvisHub makes the canvas the agent's memory

2026-07-28

An open-sourced agent runtime replaces the chat transcript with a typed canvas graph storing artifacts, versions, dependencies and provenance, so an agent can point at a specific rejected draft instead of re-reading its own conversation.

agents · agent-memory · open-source · tool-use · research

StateAct: agents that edit the file instead of the screenshot

2026-07-28

A new agent design gives computer-use agents direct code access to the files, databases and DOM behind an application instead of making them work from screenshots, reporting about a third more completed long-horizon tasks at roughly a ninth of the cost.

agents · computer-use · research · tool-use · benchmarks

Chinese open models passed US models in OpenRouter token share

2026-07-28

OpenRouter's own usage data shows Chinese models overtaking US models in token volume in early June, with DeepSeek roughly doubling its share to 18% - driven by token-hungry agent workloads routing to the cheapest capable endpoint.

open-weights · economics · deepseek · inference · agents · market

← 2026-07-272026-07-28later →