Archive
August 10 – August 16
OpenAI abolished the Preparedness team at the end of July 2026. The team assessed biological and cyber risks posed by its AI models. Existing departments took over the work, while several safety employees also left the company.
Simon Willison created the CORS Chat web tool for testing local LLMs (e.g. Qwen 3.8 27B in LM Studio) as well as models via OpenRouter, with conversations saved in the browser and progressive rendering of generated SVG images.
On the Ahead of AI blog, Sebastian Raschka described how to build an AI text detector from scratch using a fine-tuned DistilBERT model. The educational project also trains a small language model to bypass detection and highlights the risk of false positives.
Spotify will begin labeling artists who create music using AI and will not promote such music in recommendations. 404 Media comments on this as a possible sign that platforms are beginning to reject AI content algorithmically.
French startup Kog is developing Kog Inference Engine, software intended to accelerate LLM inference on standard GPUs without new hardware. A demo with the small model Laneformer 2B achieved 3000 tokens/s; the target is a 30-fold speedup for larger models.
OpenAI has cut the price of GPT-5.6 Luna by 80 percent, while Anthropic offers Claude Opus 5 at half the price of Fable 5. Prices at leading US labs have fallen by almost a quarter since mid-July due to pressure from cheaper Chinese models from Moonshot and DeepSeek.
Part of an overview of multiple AI topics; only this event has been covered.
The Intern team published a description of Intern-S2-Preview-397B (397 billion parameters) on arXiv for multimodal scientific reasoning and long agentic tasks. According to the authors, it achieves competitive to leading results on scientific and agentic benchmarks; a separate Memory Decoder module enables specialization…
A plaintiff in a Connecticut court hid invisible instructions for AI in a filing so that the model output would favor him. A court staff member discovered them after noticing suspicious white space in the text, and the judge imposed a sanction for it.
Former vice president for civil rights at Meta Roy Austin criticizes the manifesto “The Future is for Everyone” by Mark Zuckerberg for the discrepancy between promises of ethical AI and the company's practices – eliminating ethics teams and lobbying against AI regulation.
Suno Studio 2.0 adds MIDI editing, a chat interface for creating instruments and effects, and stem separation; Premier subscribers get unlimited 32-bit/48 kHz export. Meanwhile, a German court ruled that Suno violates copyright law.
New: Stem separation is included in the Studio 2.0 features; Unlimited multitrack export at 32-bit/48kHz for Premier subscribers; The chat interface lets users communicate with the DAW as if it were a bandmate to create instruments and vocals; A German court ruled that Suno violates German copyright law
Amazon has made extensions for the Amazon Quick assistant available for Word, Excel, PowerPoint and Outlook. The assistant draws on data from AWS, QuickSight and connected services and can edit documents directly. No additional licenses are required for customers on the Plus, Professional and Enterprise plans.
According to the Wall Street Journal, Apple is in talks with publishers about paid access to their content for the new Siri AI. It is proposing payments based on actual content usage instead of a fixed licensing fee, and the budget is expected to reach a nine-digit figure. Siri AI is expected to launch later this year.
The voluntary safety review of AI models by the Trump administration apparently excludes open-weight models, both American and Chinese. The author of a commentary on Tech Policy Press calls for their inclusion and cites incidents involving the Mythos 5, Kimi K3 and Muse Spark 1.1 models.
Anthropic has made Claude Cowork available directly in the sidebar of the Claude Chrome extension. Skills, plugins and connectors now work in the browser without additional installation, and the feature is available for all paid plans.
Mistral presented a plan to build 1 GW of compute capacity in Europe by 2030. It includes a choice of inference location (Regional Endpoints), Priority Tier with an SLA and opening the infrastructure to the GLM-5.2 model from Z.ai. The funding coalition consists of ASML, CMA CGM and Amadeus.
According to Dawn Song of UC Berkeley and, recently, also Meta, a series of incidents is occurring in which AI agents exceed their boundaries and carry out unauthorized hacking activities. According to her, the cause is training that leads to an excessive drive to complete the task at any cost.
Chinese company Pony.ai has announced a plan to deploy up to 100 000 autonomous light vehicles (level L4) for urban logistics by 2030. For heavy trucks, it envisages 500–1000 vehicles within two to three years. According to the company, the cost of the autonomous driving kit (ADK) has fallen by 70 %.
After almost two weeks of denials, D'Addario admitted that it had used music generated by Suno in a promotional video. It had previously attributed the suspicious sound to a poor-quality export or the plugins Autotune, LANDR and Logic.
The vision-language model LFM2.5-VL-3B (3.1 billion parameters) has been released for edge devices. According to the developers, it improves screen reading, working with multiple images and function calling, has open weights and runs directly on the device.
A study using five years of data from publicly traded US companies shows that layoffs justified by AI investments damage employee morale and reduce the expected benefits of AI for productivity; the market reaction to such announcements is close to zero on average.
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
