Five MIT faculty members and two alumni have been named to the AI2050 Fellows 2026 program. The initiative, funded by Schmidt Sciences, focuses on tackling hard problems in artificial intelligence through research projects centered on AI and society.
An agent running on the Claude Sonnet 5 model from the startup Exin Therapeutics independently selected a gene therapy and performed brain surgery on a mouse with epilepsy; midway through the procedure it refused to continue after detecting a calibration error. The results have not been peer-reviewed.
An employee of NVIDIA created the MonsterHouse Halloween installation – a talking garage door playing the role of a grumpy pirate ship. It uses the Nemotron Voice Agent pipeline with the Nemotron ASR, Bolt E4B, Magpie TTS, and Audio2Face-3D models, running offline on a single workstation with RTX PRO cards.
Startup Zest has launched an AI agent available via SMS (number 212-470-6067) that recommends restaurants based on a credit card linked through Plaid and data from friends. The agent books tables via OpenTable, Resy, and SevenRooms and also works in group chats with voting. Zest was founded in November…
OpenAI has restructured the Model Spec behavior guidelines around five values and tightened requirements for the correctability of agents, which must not block oversight or shutdown. This follows the previously described multi-layered monitoring of undesirable model behavior.
Starbucks has announced its intention to explore AI technology for drive-thru that would recognize logged-in customers and display personalized menus to them. Similar systems are already being tested by Taco Bell and McDonald's.
After the Black October wave of vulnerabilities (PHP2Shell, GitLab, the Binance incident), Nathan Lambert of AI2 maintains that open weights for the Claude Mythos model would not have been a catastrophe – in his view, that threshold was crossed by GLM-5.3. This builds on his earlier criticism of the debate over banning open models.
The company Worldwide Express deployed a multi-agent AI solution on the Amazon Bedrock platform to automate freight invoice disputes. According to the company, this reduced active exceptions by 75% and freed up $10 million in cash flow.
The MVP test satellite from Google's Suncatcher project has been communicating with the ground since its launch on 1 October 2026. The Gemma 4 model on board generated a haiku about being in space; due to power constraints, only one of the four TPUs can be switched on at a time, for about 20 minutes.
Following an article by editor Jason from 404 Media about ChatGPT-generated tech support, he noticed that he is receiving AI-emails from people whom ChatGPT recommended to contact him. The model generated instructions with his real name and contact details. The phenomenon also affects other journalists and experts.
Pope Leo XIV called on the UN to take concrete steps at the meeting in Geneva to ban lethal autonomous weapons, emphasizing the need for human oversight over AI decisions.
Google DeepMind is piloting the PlanAssist tool at British building authorities (Barnet, Camden, Dorset since June). The application validates documents, checks compliance with policy, and produces reports with justification. The pilot showed: assessment time cut in half, 70% of reports approved by officials without changes.
DoorDash disclosed that third-party AI agents (Muse, Instinct, Fo) have generated only a few hundred orders over their entire operating period, compared to nearly 1 billion total orders in Q3 2026. CEO Tony Xu called this very low usage.
CEO Sam Altman confirmed that OpenAI's leadership has more mathematical proofs in store than have been published so far. Altman defends the public's right to scientific knowledge, even though mathematicians criticize the poor communication in releasing hundreds of results.
Runware plans serial production of modular Sonic Inference Pod data centers at a factory in Romania, aiming to reach a capacity of 30 MW per day by 2028. A second factory is to be launched at the beginning of next year, and pod installation is expected to take 48 hours.
A commentary on The Conversation warns that using ChatGPT for flirting or dating communication can deprive a person of self-awareness of their own emotions and carries risks of privacy violations and inaccurate representation of the person.
OpenAI will respect a shared signal in robots.txt that allows website operators to express preferences regarding the use of content for AI training. The signal is meant to apply across multiple AI providers.
OpenAI has published a safety report with expanded protective measures in the ChatGPT product aimed at preventing violence and protecting teenagers, including a commitment to regular public safety reporting.
NVIDIA has published a technical description and benchmark results for the Nemotron 3.5 Super (120B/12B active) and Lightning (30B/3B active) models. The Lightning model itself was released by the company as open source under the OpenMDW 1.1 license back on 11 August, together with the NeMo Switchyard routing tool.
A study on the Mistral-7B and Llama-3-8B models compared two approaches to maintaining safety alignment during sequential LoRA fine-tuning: supervised replay and logit distillation. Distillation achieves a smaller reduction in attacks, but damages the model's willingness to answer harmless queries far less.
The new Jailbreak Connectivity method achieves, according to the authors, an 84–98% success rate at bypassing the safety guardrails of multimodal language models on the MM-SafetyBench benchmark, including black-box tests.
The University of Turin published a textbook-style book on LLMs on arXiv for scientists and professionals without programming knowledge, including a practical exercise on local deployment using Ollama and Open WebUI.
David Durman, an administrative employee at the University of Kentucky, obtained an exemption from using AI on the basis of his humanist beliefs. The university said it assesses requests for religious accommodations in the workplace on a case-by-case basis. The available excerpt does not specify the scope of the exemption.
On 30 October 2026, OpenAI released the GPT-6 Rosalind model for biological reasoning and scientific tools in life sciences research. Access is available only to organizations approved under the Rosalind Research program.
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.