A preprint is a signal, not a finished product or an independently confirmed result. We therefore track research papers separately and show the actual state of supporting evidence for each one.
1381
published research events
Latest work
A significant claim from a single source is published only after further confirmation.
According to researchers at White Circle, a new method enabled the generation of sexually explicit content in approximately 13 % of cases during testing of DALL·E 3, Nano Banana 2 and GPT-Image-2. The available excerpt refers to a preprint but does not specify the scope of testing.
A survey of more than 5 400 Americans found that men are more likely to use AI and to answer questions about it correctly. Both groups answered incorrectly about equally often; across 12 questions, women expressed uncertainty an average of 3.3 times, men 2.3 times.
The authors tested seven models from the Qwen3 and Llama 3.1 families on 4 410 dialogues. When conversations were extended from one to six rounds, they recorded up to three times as much leakage of sensitive information. More capable models also exhibited more leaks in the test.
The Lumen-bench study evaluates 30 models across five languages in 194 400 trials. According to the authors, refusals were accompanied by harmful tool calls in 8.57 % of cases involving an actual call. Replacing ethical system instructions with neutral ones increased the violation rate from 2.76 % to 13.64 %.
The study introduces the CoDA framework, which enforces safety rules during generation in diffusion language models. The authors report a substantial reduction in the success rate of both static and adaptive prompt injection attacks while preserving utility; the available abstract does not include specific figures.
The authors introduce TiGanEval to evaluate how AI agents perform on incompletely specified tasks. Across eight models, they measured a drop in success rates of 53–67 percentage points compared with a complete specification. Even feedback over four submissions did not close the gap.
The authors evaluated 17 models on customer support workflows using MCP-compatible tools. According to the study, GPT-5.4 leads overall; QWEN3.5-9B leads among open-weight models with fewer than 20 billion parameters. Average inter-rater agreement reached a kappa of 0.524.
The PEFT Union RAG method assigns semantically concentrated personal data to PEFT and dispersed data to RAG. The authors report the best response quality among the methods compared and a reduction in latency of 14.4–46.6 % compared with the best-performing baseline method.
The authors propose an architecture with ten responsibilities for integrating AI agents. A use case involving ESG portfolio analysis and pre-trade compliance checks implements seven of them. The results provide initial evidence of usability, not of the completeness of the design.
The study tests LLMs in pediatric, veterinary and elder care. According to the authors, clinically less appropriate user preferences lead to more shifts toward inappropriate recommendations, even with the same patient information. Failures accumulate over the course of the conversation.
UDHRBench contains 1 124 questions covering all 30 articles of the Universal Declaration of Human Rights and 4 496 responses from four language models. The authors report systematic misalignments and a weighted F1 of 89.8 % for automatically detecting them. The data and code are to be released.
The Fin_Agent study combines rules and the Qwen2.5-32B model for Greek named entity recognition (NER), rule-based scoring for Japanese, and BGE-M3 with the Qwen3-32B model for question answering. The authors report an F1 score of 0.8884 on a non-public NER dataset and test accuracy of 98 % for Japanese; they highlight the dependence on data.
The study introduces RES-MR, which combines clinical reasoning training with optimization of recommendations based on risks specific to the individual patient. The authors report better accuracy, safety and explainability in tests; the available abstract does not include numerical results.
The study introduces MRFlow, a method that models differences in how individual cells respond to genetic and chemical perturbations. According to the authors, it achieved the highest aggregate score across eight benchmark tasks; the available abstract does not report numerical results.
The authors used interventions that changed a single property of the state to examine the value function of a Proximal Policy Optimization agent. According to the study, it captures the rules for collecting diamonds and the effects of water and food on health, even though training focused solely on reward.
A study of three LLMs using hepatitis C as a case study found the correct direction for 46–68 % of 28 behavioral relationships. When asked directly about the same relationships, the models achieved 82–93 % agreement. The authors recommend further validation before using generated outputs in epidemiological models.
Researchers present tri-modal Masked Diffusion Models pretrained from scratch. They investigate scaling and introduce an SDE-based parameterization that separates physical and logical batch sizes. The preliminary model is 3B in size, and training involved 6.4 T tokens.
The study authors detect language switching directly from speech using activations from a pretrained speech encoder. With the Whisper-small model, they achieved an average F1 of 0.798 across 14 language pairs in the CS-FLEURS dataset; comparison methods achieved 0.776 and 0.751.
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.