Archive · month
Startup Kepler Computing has emerged from stealth mode after seven years of development, claiming that its 3D stacking approach to increasing HBM and SRAM memory density does not require EUV lithography. The company has raised 468 million USD from investors and received a pledge of up to 245 million USD from the US Department of Commerce.
Startup Cymphony has raised 30 million dollars in funding (including 25 million in Series A) from Sequoia Capital and SMBC Fin Atlas Beyond Fund for a platform that monitors AI agent access to company data. Its post-investment valuation exceeds 100 million dollars.
Amazon Prime Video has launched a feature that uses AI and VFX to synchronize actors' lip movements with dubbed audio. Currently available only for the English dub of the German series Maxton Hall, the company plans to expand it to additional titles.
AWS and Qualcomm are collaborating on the development of custom AI inference chips and optical interconnects with a capacity of 1.6 Tb/s. Qualcomm uses AWS Bedrock for chip design; this is its third major data center contract since June, with a goal of 15 billion USD in revenue by 2029.
GitHub has launched agentic autofix in Code Quality: Copilot can fix up to 25 findings at once, verify its changes itself, and open a pull request for review. Available on GitHub Team and Enterprise Cloud, it consumes AI credits.
Mistral AI helped a European energy company migrate a reservoir simulator containing 40 000 lines of Fortran 77 to C++ using AI agents. The key was a verification harness for numerical parity and a structured team of agents with human supervision.
An essay by Nathan Lambert (AI2, Interconnects blog) argues that, unlike the industrial revolutions, AI has yet to bring tangible benefits to ordinary people, and there is a risk of political opposition due to the uneven distribution of benefits.
Mathematician Terence Tao warns that even a rumor of work on an open scientific problem now triggers a massive AI effort that outpaces the original research – he says this threatens to end the tradition of openly sharing scientific research directions.
New York has banned student-facing AI in grades 1–8 starting in the new school year, allowing high school students to use only a limited list of tools. Los Angeles Unified School District announced a one-year moratorium on generative AI for approximately 378 000 students on the same day.
Google has released WeatherNext v3, a weather forecasting model that now processes satellite data, shortening the delay between current weather conditions and a new forecast. According to the company, it achieves accuracy comparable to traditional models with lower computational requirements.
AWS described a GitHub Actions pipeline that deploys an agent to AWS Bedrock AgentCore, evaluates it using AgentCore Evaluate API, and blocks a pull request when quality deteriorates; it also addresses OAuth authentication for CI against an MCP server with roles.
DiDi International Business Group collaborated with AWS to build an in-house QA system for contact centers on Amazon Bedrock with three pipelines (intent verification, compliance, VOC analysis); intent verification accuracy increased from 38 % to 86 %.
OpenAI resolved an incident involving elevated error rates in image generation. According to the company, the affected Image Generation and Images services have fully recovered and are operational again.
Danijar Hafner, a former researcher at Google Brain and Google DeepMind, left Google DeepMind in autumn 2025 and founded a stealth startup in San Francisco. He is developing AI agents using model-based reinforcement learning and world models, tested on humanoid robots imported from China.
Commentator Simon Willison tested the new image models from OpenAI, GPT Image 2.5 Sunburst and Flare, recommends Sunburst for precise edits and Flare for fast generation, and also verified editing using a reference image.
New: The model supports image-to-image generation with a reference image for editing; Simon Willison (an authoritative LLM commentator) tested both models and recommends Sunburst for precise edits, Flare for speed; Practical verification: the CLI tool integrates reference image support for image generation
According to its own claims, 1Password increased engineer productivity by 21 percent by using Codex from OpenAI to develop new features and internal tools.
Košík.cz has become the first retailer in Central Europe to launch a plugin for shopping through ChatGPT. According to CEO Ivan Utěšil, it currently generates only a few orders a day, but the company expects 15 % of revenue to come from this channel by 2030.
The government of South Korea has launched the “AI for Everyone” program, which is intended to give citizens free, unlimited access to AI in public services as standard infrastructure.
Eric Provencher from OpenAI issued recommendations on 12. 9. 2026 on how to shorten skill descriptions, limit mandatory documentation reading in AGENTS.md and reconsider guardrails for the GPT-6 Astra model—according to the company, accumulated rules unnecessarily take up context and stop the model's work prematurely.
New: Eric Provencher from OpenAI issued specific recommendations for simplifying prompts and guardrails; Shorter and more specific prompts improve the performance of the GPT-6 Astra model; Unnecessary guardrails and blanket rules hamper performance; Skills are Markdown files containing instructions that are selected based on their descriptions; Long skill descriptions lead to poor selections and context for you
Xiaomi brought the humanoid robot CyberOne to Europe. According to the company, it makes autonomous decisions and gains experience while working in an electric vehicle factory. You can find details in the source article.
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
