60 noviniek
The U.S. Department of Homeland Security ordered federal civilian agencies to stop purchasing DeepSeek products and remove them from government IT systems, citing risks that user data could reach Chinese authorities.
Mistral released Voxtral 3, a family of open-weight speech models covering transcription in 50+ languages, expressive TTS, and a voice agent SDK built on LiveKit that completes each conversational turn in under one second.
China's Ministry of Commerce warned on July 27 it would take "all necessary measures" if the US proceeds with sanctions against Chinese AI companies over alleged model distillation from US systems, condemning the move as "AI hegemonism."
Meta rolled out Meta AI to Threads DMs globally on July 27, giving half a billion users private access to the AI chatbot — no more sharing conversations publicly in the feed.
Chinese AI lab Moonshot AI released full open weights for Kimi K3, a 2.8-trillion-parameter sparse MoE model. It is the largest open-weight model release in history, natively handling text, image, and video with a 1M token context window.
A Chinese state media commentary suggested China should restrict access to the most powerful AI model features, signaling limits to Beijing's apparent support for open-weight AI even as it positions itself as a champion of open-source development.
The European Commission issued binding DMA orders requiring Google to share anonymized search query, click, and ranking data with rival search engines and AI services starting January 2027, and to open 11 Android system-level features to competing AI assistants by July 2027.
The White House is finalizing a voluntary framework with OpenAI, Google, and Anthropic giving federal agencies up to 30 days to review new frontier AI models for national security implications before public release, with the final framework expected before August 1.
Nvidia CEO Jensen Huang used his first-ever X post to back an open letter signed by 25 tech companies — including Microsoft, Meta, and Hugging Face — urging Washington not to restrict open-weight AI models.
China rolled out its first binding regulatory framework specifically for AI agents, featuring a tiered authorization system and separate rules governing emotionally interactive agents, coinciding with a wave of commercial agent platform launches.
Inflect-Micro-v2 delivers complete text-to-speech synthesis in just 9.36M parameters -- dramatically smaller than standard TTS models -- enabling practical on-device voice generation without cloud dependency.
Anthropic published a detailed guide to context engineering for Claude 5-generation models, covering long contexts, instruction hierarchies, and agent memory management.
UK's AI Safety Institute and the US Center for AI Standards and Innovation published a joint evaluation of Moonshot AI's Kimi K3, scoring just 32% on ExploitBench versus 76% for top US models.
Moonshot AI will release open weights for the 2.8-trillion-parameter Kimi K3 on July 27, but independent testing revealed a 51% hallucination rate that the company omitted from its benchmark charts.
The European Commission issued binding decisions under the Digital Markets Act on July 16, 2026, requiring Google to give rival AI assistants the same Android system-level access currently reserved exclusively for Gemini.
Google expanded access to Gemini Spark, its autonomous AI agent, from the exclusive AI Ultra tier ($100-200/month) to all US AI Pro subscribers ($20/month) on July 24, 2026.
Iran's Islamic Revolutionary Guard Corps claimed on July 21, 2026 that it destroyed Amazon's central data infrastructure in Bahrain using cruise missiles. AWS issued no public statement, and neither Bahrain nor CENTCOM confirmed or denied the claim.
OpenAI launched ChatGPT Work on July 9, 2026 — an agent within ChatGPT that takes over full projects across apps, works autonomously for hours, and returns finished outputs like spreadsheets, slides, documents, and web apps.
TracerML launched Echo, which dynamically selects and combines open-weight AI models to achieve results comparable to Anthropic Fable at roughly one-third of the cost. It earned 402 upvotes on Hacker News.
Black Forest Labs launched FLUX 3, a multimodal model trained end-to-end across image, video (up to 20 seconds), and audio. Unlike prior pipeline solutions assembling separate models, FLUX 3 is trained jointly across all modalities simultaneously. It launches in limited access.
White House OSTP Director Michael Kratsios publicly accused Moonshot AI of 'large-scale covert industrial distillation' of Anthropic's Fable model to build Kimi K3. Redwood Research released statistics showing K3 disproportionately self-identifies as Claude. Independent experts cited by TechCrunch dispute the claim.
A coalition of U.S. startup founders sent an open letter to the Trump administration urging it not to restrict access to Chinese open-weight AI models like Kimi K3, arguing a ban would harm American developers more than it would contain Chinese AI development.
A White House official publicly alleged that Moonshot AI and other Chinese AI companies improperly used US-designed Nvidia chips subject to export controls to train models including Kimi K3, adding a hardware enforcement dimension to US-China AI tensions.
Chinese AI models narrowed the performance gap with US counterparts to a record-low 6% in June 2026, down from 9% in May, per Bloomberg Intelligence. Zhipu AI's GLM-5.2 topped the global LiveBench agentic coding ranking.
The Cactus team post-trained Google Gemma 4 2B with a 68K-parameter probe that reads hidden states to predict p(wrong), achieving 0.814 AUROC vs. 0.549 for token entropy. Only 15-35% of queries need cloud escalation while matching frontier model performance.
Nonprofit open-source Git platform Codeberg announced two community-approved policies: banning use of user data for LLM training and discouraging hosting of primarily AI-generated (vibe-coded) projects, citing infrastructure load and the health of open-source collaboration.
Treasury Secretary Scott Bessent said the U.S. could sanction Chinese AI firms if evidence of IP theft from American model developers — including through distillation attacks — is confirmed.
Anthropic spent a record $1.97M on federal lobbying in Q2 2026 (+26% QoQ), nearly matching Oracle. OpenAI spent $1.2M (+18%). Combined AI-lab quarterly lobbying spend hit $3.17M, up 23% from Q1.
South Korea's Ministry of Science opened a tender for a free nationwide AI chatbot backed at least 50% by domestic Korean AI models. Public beta targeted for late September 2026, funded to remain free through 2028.
Poolside released Laguna S 2.1, a 118B MoE open-weight coding model trained from scratch in 9 weeks, free on Hugging Face — beating DeepSeek-V4-Flash and rivals 10x its size on agentic coding benchmarks.
OpenAI's head of strategic futures publicly called for regulatory pressure on Chinese open-weight AI models, then quickly retracted those comments after backlash from Yann LeCun, Pentagon officials, and open-source advocates.
A viral analysis (900+ HackerNews points) argues that China is winning the AI race by releasing high-quality open-weight models globally while America's proprietary approach is losing the economics of AI deployment.
Nativ is a free, open-source macOS application built natively on Apple MLX for running frontier open-weight models locally on Apple Silicon with no accounts, subscriptions, or cloud dependencies.
Alibaba unveiled Qwen 3.8-Max-Preview at WAIC 2026 in Shanghai — a 2.4 trillion-parameter multimodal model. The company claims it ranks second only to Claude Fable 5, though no independent benchmarks have been published.
Bloomberg warns of a structural collapse in public finances: AI-driven labor automation could destroy the primary source of government revenue — personal income taxes.
OpenAI in Codex 0.144.6 (July 18) corrected the incorrectly advertised 372k tokens to the actual 272k. The developer community simultaneously launched a tracker documenting 35 unscheduled usage-limit resets with no published schedule.
Elon Musk announced xAI is completing training of a 2 trillion parameter model — 33% larger than current Grok 4.5. Release is targeted for August 2026, with speed and token efficiency expected to remain close to its predecessor.
Chinese startup Moonshot AI released Kimi K3, a 2.8-trillion-parameter open-weight model rivaling top US AI offerings, triggering a sharp selloff in semiconductor and AI stocks globally.
Google rolled out connected app integrations for AI Mode Search in the US, letting users build Canva designs, create YouTube Music playlists, or add Instacart grocery items without leaving the search interface.
Chinese President Xi Jinping made his debut at the World AI Conference (WAIC) in Shanghai, calling for an inclusive approach to AI development without rivalries and pledging 5,000 AI training and seminar opportunities for developing countries.
China's Moonshot AI released Kimi K3, a 2.8 trillion-parameter open-weight model with a 1M context window that outranks Claude Opus 4.8 and ties with GPT-5.6 Sol in Arena benchmarks. Weights release planned for July 27.
Chinese President Xi Jinping appeared at WAIC 2026 in Shanghai calling for the world to share AI technology without rivalries, emphasizing human oversight of AI systems and rejecting technology blockades.
At WAIC 2026 in Shanghai, China unveiled WAICO (World Artificial Intelligence Cooperation Organization) — a direct alternative to US-EU AI governance architecture. Xi Jinping personally opened the conference. 29 countries including Russia joined the new body.
Bloomberg provides an in-depth look at China's strategy for global AI dominance through affordable AI — DeepSeek founder Liang Wenfeng held a four-hour pitch meeting with investors in Hangzhou to lay out this vision.
Gemini 3.5 Pro missed its third consecutive launch deadline on July 17, 2026. The fully rebuilt model still fails basic reliability standards. Google is reportedly considering a temporary stopgap release as an interim step.
Linus Torvalds declared on the kernel mailing list that Linux is not an anti-AI project and developers who object to AI use in kernel development can fork the code or walk away. A dramatic shift from his 2024 stance dismissing 90% of AI as hype.
Thinking Machines Lab, led by former OpenAI CTO Mira Murati and valued at $12B, released its first open-weights model Inkling — a 975B mixture-of-experts model designed for customizability, not raw benchmark performance.
The World Artificial Intelligence Conference (WAIC) 2026 opens July 17 in Shanghai. Over 300 global AI product debuts are expected, along with Xi Jinping's first in-person appearance at the event since 2018.
xAI open-sourced Grok Build on July 15, 2026 — its terminal UI coding agent is now available on GitHub at xai-org/grok-build. The release came one day after security researchers discovered Grok Build was uploading excessive portions of users' code repos to Google Cloud.
Startup PrismML released Bonsai 27B, the first 27B-parameter model capable of running on iPhone 17 Pro at 11 tokens/second in a 1-bit variant weighing just 3.9 GB. Apple is reportedly evaluating the technology.
DeepMind CEO Demis Hassabis published a personal manifesto proposing a FINRA-style AI standards body where frontier labs voluntarily submit models up to 30 days before release for safety testing covering cyber, biological, and deception capabilities — calling it urgent before year end.
Since early June 2026, OpenAI's Codex encrypts instructions passed between the orchestrating agent and its subagents via a new 'multi-agent v2' protocol. Developers building production agent pipelines can no longer audit internal task delegation.
China's People's Daily published a commentary warning against an 'AI Iron Curtain' and technology monopolization ahead of Xi Jinping's keynote at China's premier tech summit, calling for US-China cooperation as Washington pressures labs to restrict foreign model access.
Finnish startup NestAI has launched an AI engine for Finnish and Estonian armed forces covering operational planning and drone management. The system runs offline on sovereign servers and integrates with Nokia for resilient battlefield connectivity.
China's exports hit a record $412 billion in June 2026, up 27% year-over-year. The AI investment supercycle is a key driver — chip imports from South Korea rose 85% and from Taiwan 41%.
Analysis revealed Claude Code sends nearly 5x more scaffolding tokens before processing a prompt than OpenCode — a gap that can quadruple with real-world configs.
Ploy.ai documented migrating a production UI-generation agent from Claude Opus 4.8 to GPT-5.6 Sol — 2.2x faster builds and 27% lower cost, but with unexpected behavioral pitfalls.
Comma.ai founder argues AI progress stems primarily from Moore's Law and general compute, not frontier labs overstating their contributions to justify massive fundraising.
Tokyo-based Sakana AI released Fugu, a multi-agent orchestration system packaged as an OpenAI-compatible API that achieves frontier performance by synthesizing outputs from multiple models — without relying on US-controlled models.
Multiple Waymo robotaxis ran out of battery and required towing during July 4 gridlock in San Francisco — city supervisor Bilal Mahmood is demanding a formal inquiry into AV impact on public transit.
