AI Flash · live feed
A timeline of everything that has passed through the radar — newest first, with an evidence status for each item.
Continuously updated
1142 items in the selected feed
How to read evidence statuses ↓
NVIDIA is expanding Agent Toolkit with PhysicsNeMo and CUDA-X libraries for autonomous engineers with physics simulation capabilities. The Nemotron 3 Ultra model leads among open models in RTL coding.
The article covers fraud involving Správa nemovitostí Hradec Králové (CZK 46 million) through social engineering, and the energy demands of data centers for large language models. The second part, about Zhejiang University research, is truncated.
The debate around Kimi from Moonshot AI concerns US worries about competitiveness. OpenAI and Anthropic are reportedly lobbying regulators over concerns about open Chinese models. Participants in the discussion highlight overreactions and the possibility that regulation could benefit only large companies.
The article describes a marketplace for discounted proxies to LLM APIs, primarily in China, where resellers circumvent restrictions and reduce costs by abusing free trials and sometimes stolen payment cards. It highlights the risk to developers exposing LLM applications and suggests more precise limits on API keys.
According to the New York Times, the Trump administration favors targeted bans on specific Chinese models rather than a blanket ban. OpenAI and Google DeepMind signed a letter opposing regulation of open models; Anthropic and OpenAI are lobbying for restrictions on them. Chinese models lag behind in safety tests.
Anthropic reports increased request error rates for Claude Fable 5, Claude Sonnet 5, Claude Haiku 4.5 and other models on the status page. claude.ai, Claude API, Claude Code and Claude Cowork are affected, and the cause is being investigated.
Anthropic reported increased error rates on its status page for the models Claude Mythos 5, Fable 5 and Haiku 4.5 (the title also mentions Opus 5). claude.ai, API, Claude Code and Claude Cowork were affected; a fix was being implemented.
The Anthropic API experienced temporarily elevated error rates for Sonnet 4.6 and Sonnet 5 for approximately 10 minutes. The incident was resolved around 11:07 PDT.
Librarians in the US are holding workshops on turning off AI features, especially Apple Intelligence and Gemini. Hannah Cyrus from Maine published an article about her own workshop, inspiring dozens of librarians around the world. The workshops drew record interest — attendance at the first two sessions was around…
Seven GPT models (GPT-5.2, GPT-5.3-Codex, GPT-5.4, GPT-5.4 Mini, GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna) experienced reduced availability in Copilot products and IDEs. The cause was an upstream provider failure. The incident was resolved and the service is being monitored.
NAVER, NVIDIA and Brookfield plan to expand AI infrastructure in South Korea from 55 megawatts to 200 megawatts by 2028 on the NVIDIA DSX platform. NVIDIA is investing 1 billion dollars, while Brookfield pledges up to 9 billion. The infrastructure will be located in the GAK Sejong data center and serve companies in Korea…
A technical report on the Naver-News-KO dataset with 27 400 (document, summary) pairs from Korean news from July 2022, available on Hugging Face Hub since January 2023. In May 2026, it reaches about 33 000 downloads per month. It includes baseline models (KoBART, Gemma-2B-ko with LoRA) with a ROUGE-1 score of 56.6 % and…
OpenAI released the Micro keypad ($230) in collaboration with Work Louder. It has 6 configurable keys for ChatGPT and the Codex agentic coding tool, and supports Bluetooth and USB. Developer reviews are critical — citing a high price and questionable practical usefulness compared with DIY alternatives.
Prentis, a startup co-founded by Reid Hoffman and Marc Pincus, is raising $100 million at a $1 billion valuation. It focuses on AI agents that automate office processes. It already has contracts worth up to $50 million. The company says its Hive-32B model outperforms competitors on benchmarks.
New Brunswick legislator Bill Oliver exposed a language-model instruction during a speech by reading the text "here is a more natural version of this section that sounds like a parliamentary speech...". The video spread on social media, and Canadian media called it a symbol of growing reliance on AI without review.
Cognition, the startup behind the Devin coding agent, acquired AI assistant Poke for a low nine-figure amount. Poke exchanged 100+ million messages within three months. Cognition wants to integrate Poke's friendly communication style into Devin; Poke itself will use Cognition's infrastructure.
The podcast The Vergecast discusses the Google Zero phenomenon—the decline in website traffic from Google search due to AI answers. Reddit is reportedly considering withdrawing from the training data agreement, while publishers are considering blocking the Google crawler.
Anthropic has released Claude Opus 5 at the same price as Opus 4.8 (5 USD/25 USD per million tokens), which the company says is half the price of Fable 5. It leads in agentic coding and ARC-AGI-3, while lagging in cyber exploitation and some domains such as healthcare and law.
Anthropic introduced Opus 5 – according to the company, it is cheaper and less restricted than Fable 5, and even outperforms it on a number of benchmarks in its own announcement. The new Automatic Fallbacks beta feature redirects queries blocked by safety filters to a less capable model instead of returning an error message.
Meta is adding new features to its AI chatbot — calendar access for event planning, daily briefing generation and deeper research. The upgrade is powered by the newly released Muse Spark 1.1 model.
Microsoft signed a letter supporting open-weight models, but the text says the primary aim is to strengthen Azure and reduce costs. It is also replacing OpenAI and Anthropic models in Copilots with its own MAI models, which are weaker according to benchmarks.
AWS describes the architecture of a Next-Best-Product recommendation system for banks. It uses PyTorch and Amazon SageMaker with a multi-tower neural network and a learned attention mechanism to improve accuracy and interpretability, which matters for regulatory compliance. The article includes an architectural overview and…
Bluesky now offers Quests in the AI assistant Attie, a feature that allows users to ask questions and search for information on the social network, including trends and influential accounts. Attie, launched in March 2026, is available in beta via a waitlist and is free for now; Bluesky is gradually developing it as a tool for…
Large AI companies (Anthropic, OpenAI) lobby for regulation of Chinese models and distillation; smaller startups (Little Tech Association) oppose this, arguing for market freedom. Investors are also divided — Bill Gurley supports the free market, while Chamath Palihapitiya criticizes the protection of corporate interests.
AI Radar monitors Czech and international sources every day, looking for changes that truly deserve attention.
The result is a quick overview of what has actually changed in the AI world, rather than another stream of articles.
Use the CS/EN switch to read the same Radar in Czech or English. English content is published after its translation has been checked, so new and older items may appear later.
Each row represents one event — not one article. At a glance, you can see its significance, credibility and main point.
Illustrative example, not a current news item.
Agent mode is available to all paying users
Until now, the mode was available only on the highest plan; it is now available on all paid tiers without a waitlist.
The detail page contains a fuller summary, its significance and original sources. Practical impact appears in the detail and the For individuals and For businesses views. An AI Flash item reaches the main selection only after it has been expanded and meets the publication rules.
We first summarize each event in the same way for everyone. Based on those same facts, we then explain what the change means for your own use and what it could mean for how a company operates.
Today’s briefing is the same for everyone. Pages For individuals and For businesses can be found in the main navigation — they select only events relevant to the given use case.
Reset history · Outage history · Subscribe in a feed reader (RSS) · Methodology and label explanations
