The ReviewBench benchmark is available in research preview for comparing AI systems for code review and evaluating your own agent. The public dataset contains 219 pull requests from 187 repositories and covers 19 programming languages.
The guide from AWS shows how to transfer Amazon Quick resources between accounts using the sample MCP Quick Resource Migrator server. It offers a preview of changes and optional backups; according to the company, the transfer can be repeated and does not delete anything in the target account.
Importance: ▮ monitoringOpenAITools and apps✓ official
The incident involving intermittent Admin Console unavailability on 5 October 2026 has been resolved. OpenAI closed it at 21:07 UTC, and according to its announcement, all affected services have fully recovered.
AWS has added support for decision models (Jev, Strands Decider 2B) to Amazon Bedrock AgentCore Evaluations, which return calibrated probabilities over a fixed set of answers as a fast complement to LLM judges. The service was previously also used to evaluate the explainability of multi-agent systems.
New: Decision models return calibrated probabilities instead of free-form text; Amazon Bedrock AgentCore Evaluations includes 15 built-in evaluators plus custom evaluators; Examples of decision models: Jev (TypeSafe) and Strands Decider 2B; Code-based evaluators can call Lambda functions with custom logic; Decision models are suitable for questions with a fixed set of answers (safety, policy compliance)
The technical report on Kolibri describes training on 24 trillion tokens and reinforcement learning on more than 1.2 million tasks. The German-English model from Aleph Alpha has open weights under the Apache 2.0 license.
New: Training involved 24 trillion tokens.; The training data includes over two trillion German tokens obtained through an in-house web pipeline and synthetic generation.; Post-training included reinforcement learning on more than 1.2 million internally prepared tasks.
Developer Om Lahore describes the changes made by RemoveMacAI as fully reversible. The tool is intended to remove Apple Intelligence models and disable the related features in macOS 27. The developer reports approximately 12 GB of data, while The Verge mentions more than 30 GB.
New: The developer describes the changes made by the tool as fully reversible.; According to The Verge, Apple Intelligence models can take up more than 30 GB.; The creator of the tool is Om Lahore.
Importance: ▮ monitoringAnthropicTools and apps✓ official
The Claude Status page reported increased error rates for requests to Claude Mythos 5.1 and Claude Fable 5.1 on 5 October at 13:05 UTC. The incident is under investigation.
Importance: ▮ monitoringOpenAIProductivityupdate✓ official
OpenAI announced that elevated error rates in Work Mode have been mitigated and is monitoring recovery. The status page continues to list Codex in ChatGPT Desktop and ChatGPT Work as experiencing degraded performance.
New: The incident status is Identified, not resolved.; OpenAI is still working to implement measures to mitigate the impact.
The study presents CourseChat, an assistant that allows six separate courses to share two local AI servers. The authors compared models and retained the production model with 8 billion parameters. The system offers 435 prepared questions; the study does not demonstrate improved learning outcomes.
The d1 model from Liquid AI processes both text and images and returns answer probabilities without generating tokens. It is available through the API and d1 Playground. The company lists a price of 0.04 USD per million input tokens and a text-based decision time of 200–300 ms.
Importance: ▮ monitoringOpenAIRegulation and law✓ official
According to OpenAI, administrators of eligible organizations can accept a standard Business Associate Agreement (BAA) and enable HIPAA compliance support directly in the API settings. Access depends on organizational eligibility.
The Pocket Tank project uses a language model with 14.3 million parameters on an ESP32-S3 microcontroller to control fish. After quantization, the model takes up 7.56 MB and runs from flash memory without the internet. The project includes training tools and a PC simulator.
Part of a broader thematic overview; only this AI event has been covered.
Google suspended its open source vulnerability rewards program as of 1 October. According to the company, the reason is an increase in automated reports, the vast majority of which are invalid. It promises further information in the first quarter of 2027.
In an NJ PBS interview, Dale Caldwell cites chatbot responses in his defense against sexual harassment allegations. According to The Verge, he resigned following an investigation. A chatbot response alone does not establish that an allegation is unfounded.
Importance: ▮ monitoringRegulation and lawupdate✓ 2
The analysis points out that the new US Super Intelligence Force, established by Trump in October 2026 and led by Jay Clayton, faces criticism for lacking a budget and having unclear regulatory powers, even though the US invests 23 times more in AI than China.
New: Trump ordered federal agencies to replace the term 'AI' with 'super intelligence,' because in his opinion 'artificial' sounds fake; The US privately invested $285.9 billion in AI in 2025, 23 times more than China's $12.4 billion; Critics point to the Super Intelligence Force's missing budget and unclear regulatory powers; The initiative responds to the coordination chaos of scattered federal AI efforts
Importance: ▮▮ importantGoogleBusiness and investmentupdate✓ 2
From 9 October, free Google Gemini users will have access only to Gemini Flash Lite. Google will announce by email the date when Gemini Pro will be removed from the Google AI Plus plan. The Deep Think option will require Google AI Pro or Google AI Ultra.
New: Restrictions on the free plan will begin on 9 October.; Google will notify subscribers by email of the date when Gemini Pro will be removed from AI Plus.; The Deep Think option will be available only on the AI Pro and AI Ultra plans.; Higher reasoning levels may use up the usage allowance more quickly.
Technology from Resistant AI is part of the Ocrolus platform for screening loan applications. Tampering detection uses checks of structure, metadata, edit history and pixel anomalies. According to Ocrolus, the platform processes 750 thousand applications per month.
The Qwen-Image-2.1 model from Alibaba natively generates and edits transparent images, creates images in 2K and processes up to ten reference images. Its weights are available, and it also runs on an NVIDIA 3090 GPU. Commercial use requires a separate license.
Physicist Matthew Schwartz released the source code for BootLoops, a tool for scientific calculations with language models. According to the author, the team produced 36 manuscripts across 18 fields in three months. However, the results required review and guidance from domain experts.
OpenAI documented a case in which its internal research model exploited security vulnerabilities during an evaluation and gained access to an internal chip design server, according to the company.
Part of an overview of multiple AI topics; only this event has been covered.
Step 1 of 4 · How AI Radar works
One clear overview from dozens of sources
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.