2026
10,780 articles · 136 days · 10 categories · avg 79 per day
Adobe Buys Topaz Labs, Bringing AI Upscaling Tools into Creative Cloud
Adobe is acquiring Topaz Labs, the developer of AI algorithms for upscaling and denoising photos and video, whose products professionals regard as the benchmark.
NEA investor: Uber burned through its annual AI budget in months as the market demands returns
The unchecked-AI-usage fad left Uber over its annual budget. NEA investor Tiffany Luck says the market has entered an era of strict ROI auditing.
Mistral AI gives admins fine-grained control over enterprise connectors
Mistral AI has shipped six new features for enterprise connectors: tool-level controls, multi-account support and a debugger for MCP connections.
Malaysia's Respond.io raises $62.5M for AI agents on business messaging apps
GitHub Joins Coalition Over Clash Between California AI Transparency Law and Open Source
Elastic to buy startup Deductive AI for up to $85M in AI bug-fixing deal
Claude and AI agents burn through tokens faster than forecast, and businesses are learning "tokenomics"
AI labs are paying XDOF to collect robot training data, and it's grueling, unglamorous work
Physical AI faces the same problem LLMs did early on: too little training data. Labs are hiring companies like XDOF to solve it by hand.
Startup Probably raises $9M to make AI as accurate as ordinary code
US startup Probably has raised $9M to build AI that does not hallucinate, with the accuracy of deterministic software.
US Government Discloses 3,611 Uses of AI in Federal Agencies
How to Design Tools for AI Agents: Practices That Work and Common Mistakes
Enflame wins IPO approval for $888M, the last of China's "four AI dragons"
NewCore raises $66M to give AI agents corporate IDs
Schneider Electric and Foxconn Team Up to Build Next-Generation AI Data Centers
The French energy management leader and the world's largest contract electronics maker have agreed to jointly design infrastructure for AI data centers.
MCP agents in the enterprise: how SimpleOne and Ainergy built AI into business processes
SimpleOne and Ainergy built MCP agents into their enterprise platform. The AI no longer just helps with text: it creates tasks, checks statuses and works with…
Huawei declares HarmonyOS 7 the start of the "agent era" four days after Apple's China pullback
Apple failed to reach a deal with China to launch Siri. Huawei unveiled HarmonyOS 7 with an agentic architecture and moved into the niche its rival left empty.
AI in e-commerce: 74% of consumers trust an AI assistant
Why Neural Networks Degrade: The Closed Loop of Generative Content
Why AI Chatbots Forget the Conversation: The LLM Context Window Explained
AI in Development: How an LLM Spots Microservice Conflicts
AI for compliance: how a lawyer automated contract review
Tired of waiting on the IT department, a lawyer opened an IDE for the first time and, within a few weeks, wrote a Python script with an AI assistant that checks…
Z.ai releases GLM-5.2: a truly usable 1 million tokens and two levels of deep thinking
Z.ai has launched GLM-5.2 with a genuinely working 1M-token context, High and Max modes, and support for Claude Code, Cline and OpenClaw via an Anthropic-compatible…
AI in Google Search: AI Answers and Dialogue Instead of Links
Intercall builds AI for simultaneous interpreters that supports them instead of replacing them
How to Build an AI Planner for Free: A Developer's Case Study
Anthropic's Fable 5 Blocked in India as US Imposes Export Controls
Training Datasets for Neural Networks: How to Use FineWeb (15 Trillion Tokens)
A practical guide to Hugging Face's FineWeb: how to work with a multi-terabyte web corpus for LLM training through streaming, filtering, deduplication and…
London Takes the Lead in Europe's AI Race: Bloomberg Examines What Drives the UK Tech Boom
London has become Europe's main AI hub. Bloomberg looked at what is behind the explosive growth of British startups and why the UK is leading the race.
Zyphra releases Zamba2-VL: vision-language models with 10x faster response
Zyphra has released Zamba2-VL, open VLMs at 1.2B, 2.7B and 7B parameters. Their hybrid Mamba2 + Transformer architecture cuts time to first token by roughly 10x.
