1846 articles
Microsoft is rolling out new "Projects" capabilities for Microsoft 365 Copilot, transforming the assistant from a simple chatbot into a structured project management tool that handles complex, multi-step workflows.
Alibaba just dropped its Quark AI Glasses lineup with two models—the flagship S1 and lifestyle G1. Both run on Qwen AI and pack features like text/image Q&A, near-eye navigation, and in-lens payment apps.
Fresh Similarweb data reveals ChatGPT sees steeper weekend traffic drops than Google's Gemini, pointing to different user patterns as AI tools try to expand beyond work hours.
Fresh projections show AI training demand will peak in 2025 before compute needs shift dramatically toward inference. The forecast includes growing adoption of custom chips, presenting long-term competitive challenges for NVDA.
Perplexity launched BrowseSafe and BrowseSafe-Bench as open-source tools that catch prompt-injection attacks in real time, with early testing showing the system outperforming major AI models in threat detection.
Fresh data reveals AI code-generation tools are bouncing back strongly, led by Cursor's rapid growth, while traditional automation platforms like Zapier and N8N see traffic sliding downward.
Tesla's new Optimus Gen 3 robotic hands bring dramatically improved sensitivity, dexterity, and durability to humanoid robotics. The breakthrough comes from real-world neural network training that makes movements more natural and precise.
Tesla's Optimus humanoid robot hit a new performance milestone in lab testing, with the company sharing footage of the robot running and confirming it set a new internal record.
Anthropic's internal research shows engineers now use Claude for most daily tasks, with autonomy and complexity up significantly since early 2025. Survey data from 132 technical staff reveals major productivity gains and changing work patterns.
NVIDIA introduces Alpamayo-R1, an open-source AI model that enhances autonomous vehicle reasoning by moving beyond simple object detection to context-aware understanding of real-world driving scenarios.
Mistral AI unveiled its Mistral 3 family featuring dense and MoE models trained on NVIDIA H200 GPUs. The release brings expanded multimodal and multilingual capabilities with strong benchmark performance across the lineup.
Kling O1 launched a unified multi-modal AI system that handles text-to-video generation, keyframe editing, and advanced content creation in one platform, intensifying competition in the AI video space.
Arcee released new Trinity Mini and Trinity Nano models built on a mixture-of-experts architecture, featuring efficient active-parameter configurations. Early benchmark results show competitive performance across multiple reasoning and evaluation tasks.
Simular has launched Simular 1.0, a Mac desktop agent that automates multi-step workflows using a neurosymbolic architecture combining LLM exploration with deterministic code, achieving near-human performance on industry benchmarks.
Stanford researchers have developed a breakthrough method enabling AI agents to coordinate through internal neural network signals rather than explicit communication, achieving significant performance gains and efficiency improvements across multiple benchmarks.
A new McKinsey Global Institute report finds that AI will not replace most human skills but will significantly change how they are applied. Negotiation, problem solving, and leadership are projected to gain value as automation expands.
DeepSeek v3.2 appears with a sparse attention system that cuts computational costs and still matches the performance of top AI models. This marks a clear move toward greater efficiency in the design of large language models.
Fresh Google Search Console data reveals a dramatic decline in click-through rates despite maintaining strong visibility. The numbers tell a clear story: AI-powered tools are fundamentally changing how people find and consume information online.
A new AI method called Chain-of-Visual-Thought introduces continuous visual tokens to improve reasoning in vision-language models. Recent tests show accuracy gains of 3–16 percent across major benchmarks.
Google has expanded availability of Gemini 3 and Nano Banana Pro in Search to India and the U.K., while also enabling Gemini 3 Pro for AI Pro and Ultra subscribers through AI Mode.
We use cookies to improve your experience on our site and to show you relevant advertising. To find our more, read our privacy policy and cookie policy
Eseandre Mordi
Sergey Diakov
Alex Dudov
Usman Salis
Peter Smith
Artem Voloskovets
Saad Ullah
Marina Lyubimova
