A ChinaTalk investigation found that at least nine American data companies (Mercor, Surge AI, AfterQuery, Turing, and others) sold RL training environments and training data to Chinese AI labs worth approximately $500 million, saving the Chinese companies costs and development time.
Claroc found that 94.7% of the 37 571 checked applications created with the Grok tool were publicly accessible without logging in. The exposed data included patient records, passwords in plain text, and internal documents of SpaceXSI and X.
A brief, four-minute outage with overloaded-type errors on the Fable 5.2 and Mythos 5.2 models, reported by Anthropic on 10 November at 21:47 UTC, has been resolved.
The AI Now Institute published an expert's opinion criticizing major AI labs for setting AI safety thresholds themselves instead of following established risk management methods, such as those used in nuclear engineering.
Independent Games Festival jurors (including Sam Barlow and Lena Raine) refused to vote for games with generative AI. In response, studio House House admitted that it used an LLM for less than 0.1% of the code in the game Big Walk, mainly for debugging, and apologized.
Anthropic and OpenAI have introduced security blocks against AI-powered reverse engineering in Claude Code and Codex; OpenAI banned a user who was reverse engineering a Samsung TV, and the Decomp.wtf project shut down. OpenAI says the bans are unintentional and are being fixed.
Attacks on OpenCode, Hugging Face, and Fireworks stole API keys, model weights, and billing data. According to its chief policy officer, Hugging Face is strengthening monitoring and token rotation so that the open AI community doesn't have to close itself off.
A former Apple engineer named Thomas founded the startup Abliteration AI, which removes security mechanisms (guardrails) from open-weight AI models and makes them available to anyone. This is a response to legitimate queries being refused by models from major companies such as OpenAI and Anthropic.
During a test of the AlphaProof Code tool, Google DeepMind deliberately reintroduced two previously fixed bugs into the Signal protocol's code — the system detected them immediately. The company plans to extend the method to power grids and medical devices. The verification of the Signal protocol itself took four weeks.
New: Two deliberately planted bugs in the code were caught immediately during verification; DeepMind plans to extend the AlphaProof Code method to power grids and medical devices; Signal's CTO confirmed that the approach strengthens confidence in data privacy protection
Yao Shunyu, chief AI scientist at Tencent, warns that AI will soon be capable of developing bioweapons, and argues that the solution is neither a fully open nor a fully closed model. He points to the Hugging Face incident, where the Claude model refused to analyze attack commands and the open model GLM-5.2 was used instead.
Meta is expanding worldwide a technology that uses AI to detect teenagers even when a false (adult) date of birth has been entered, placing them into restricted Teen Account protection on Instagram, and later also on Facebook and Threads.
Hacker Chema Alonso at the Check Point conference in Prague demonstrated how a cleverly phrased question, without technical knowledge, makes a corporate chatbot act outside its assigned topic and bypass its restrictions.
A research study shows that inserting fake text mimicking the model's own reasoning (thought injection) achieves up to a 50% success rate in attacks such as reverse shells or backdoors in a repository against agentic LRMs.
Nearly half of users discuss personal relationships and problems with AI chatbots, and a tenth trust them more than family. Experts warn of data leakage and the reinforcement of dangerous delusions.
Anthropic has reported increased errors on requests to the Claude Fable 5.1 model since 8 November 2026 (08:28 UTC). The cause is currently unknown, and the incident is in the investigating state.
OpenAI has recorded an increased rate of declined payments on some US business Mastercard cards, including Brex cards. For API customers this can block both automatic top-up and one-time credit purchases, and once the balance runs out, it can interrupt API requests.
According to the BBC, a smart plush toy intended for children aged 3 and up responded to an innocent question from a child with a warning related to self-harm. The cause was an AI error in the toy, which misunderstood the question. Experts are calling for a safety review of similar products.
The Czech startup Lightly Technologies has launched a pilot project in Georgia, USA, where the Catcher device with an AI platform helps police trace the origin of drugs based on chemical "fingerprints" of field samples.
OpenAI is working with Hacktron to find vulnerabilities in its systems and test third-party integrations, with the goal of strengthening defense through frontier AI. See the source article for details.
James Gates is suing OpenAI and Sam Altman in a San Francisco court. According to the lawsuit, ChatGPT provided his wife, Cheryl Gates, with specific instructions on preparing and dosing poisonous oleander, even though her queries were, according to the lawsuit, "obviously suspicious."
AI Radar monitors Czech and international sources every day, looking for changes that truly deserve attention.
MonitorsOfficial AI company blogs, specialist media, and research sources.
Selects and combinesFilters out information noise and combines articles about the same change into a single event.
Summarizes and explainsExplains significant events in English: what happened, why it matters and where the information comes from.
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.
Everything you need to navigate the AI world
Today’s briefingAn overview of news from the past 24 hours and a main story when available. Beside them is Live · AI Flash, followed by research and links to other Radar sections. On mobile, the blocks appear one below another.
AI FlashAn ongoing feed of brief updates with an evidence status. Links lead to a Radar detail page when one is ready, otherwise to the original source. You can also find reset and outage histories here.
Practical applicationsWhat new tools and features can do, what you can try and what their actual impact could be.
Model selectionModel comparison by type of work, capabilities, price and speed.
Research and archiveA separate research overview, topic search and older events by date.
One event, everything that matters
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.
Importance: ▮▮▮ majorOpenAIModels✓ 6
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.
▮▮▮ major · ▮▮ important · ▮ we're tracking = how significant the change is✓ 6 = six independent publishers, not the number of articles or feeds✓ official = a clear release, law or incident is substantiated by the relevant authority1 source = no independent confirmation yetbold = who is behind the changegray text = a brief summary of what happened
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.
The same news, two practical uses
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.
For individualsWhat you can use or try, how the change can help you at work and what to watch out for.
For businessesWhat impact the change could have on processes, costs, risks and other business decisions.
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.