Alibaba's Qwen3.5-Omni handles audio, video, images, and text in a single model pass - and generates speech in real time. The Plus variant hits SOTA on 215 benchmarks and edges out Gemini 3.1 Pro on audio tasks.
Shopify's free, MIT-licensed AI Toolkit connects Claude Code, Cursor, and Gemini CLI directly to live store data, giving agents real-time access to API schemas, documentation, and store execution.
Three papers: AI safety measures withhold critical clinical guidance from patients, SAT cuts reasoning tokens by 40%, and conformal prediction blocks wrong multi-agent consensus.
OpenAI is backing an Illinois bill that would protect AI labs from lawsuits even when their models contribute to mass casualties or billion-dollar disasters.
Intel will use its 18A process node to manufacture chips for the Tesla-SpaceX-xAI joint venture targeting 1 terawatt of AI compute annually - and claims $2B in CHIPS Act subsidies in the process.
Microsoft released the Agent Governance Toolkit, a seven-package open-source system that enforces policies on autonomous AI agents at sub-millisecond latency and covers all 10 OWASP agentic risks.
LG AI Research released EXAONE 4.5, a 33B open-weight vision-language model that posts higher STEM scores than GPT-5-mini and Claude 4.5 Sonnet - but a non-commercial license caps its real-world reach.
Intel's Arc Pro B70 launched on March 25 with 32GB GDDR6 and 367 TOPS for $949, undercutting NVIDIA's RTX Pro 4000 by $850. The hardware case is strong. The software story is not.
Researchers from Google DeepMind, Microsoft, and Columbia propose financial guardrails for AI agents, with simulations showing up to 61% reduction in user losses.
A practical, step-by-step guide to using AI tools like ChatGPT, Claude, Duolingo Max, and Talkio to learn any language faster - no prior experience needed.
A practical beginner's guide to using AI tools for fiction, stories, and creative writing without losing what makes your work yours.
A beginner's guide to using AI tools like ChatGPT and Canva to write captions, plan posts, and save time on social media.
Meta's first proprietary frontier model leads on HealthBench Hard and scientific reasoning but trails rivals in coding and agentic tasks - with no public API yet.
A deep look at Microsoft's three new in-house AI models - MAI-Transcribe-1, MAI-Voice-1, and MAI-Image-2 - and whether they live up to the hype.
Google's Gemma 4 family - four models, full Apache 2.0 licensing, and benchmark scores that challenge models 10x their size - is the most consequential open-weight release of 2026 so far.
Rankings of AI models on IFEval and IFBench, the two main benchmarks for measuring how reliably LLMs follow precise formatting, length, and content constraints.
Rankings of the most popular MCP servers across development, data, web automation, and productivity categories based on installs, search volume, and GitHub activity.
Rankings of the best AI models and agent frameworks on computer use benchmarks - OSWorld, OSWorld-Verified, and ScreenSpot-Pro - updated March 2026.
Meta's first closed-source frontier model scores 52 on the Artificial Analysis Intelligence Index, leads on HealthBench Hard, and ships free at meta.ai - but has no public API yet.
Gemma 4 is Google DeepMind's most capable open model family: four variants from 2B to 31B, Apache 2.0 license, multimodal across text/image/video/audio, and the 31B Dense ranking #3 on Chatbot Arena against all open-weight models globally.
Grok 4.20 is xAI's current flagship LLM with a 2M-token context window, native multi-agent mode, and reasoning toggle at $2.00/M input tokens.
A deep look at Microsoft's three new in-house AI models - MAI-Transcribe-1, MAI-Voice-1, and MAI-Image-2 - and whether they live up to the hype.
Z.ai's GLM-5.1 scores 58.4 on SWE-bench Pro, edging out GPT-5.4 and Claude Opus 4.6, after being trained on 100,000 Huawei Ascend chips with no US silicon.
Utah becomes the first government in the world to approve an AI system to autonomously renew psychiatric medication prescriptions, limiting it to 15 lower-risk drugs under a tightly supervised pilot.
Anthropic's restricted Claude Mythos Preview model autonomously discovered thousands of high-severity vulnerabilities across every major OS and browser, including bugs hiding in plain sight for 27 years.
Project Glasswing unites AWS, Apple, Google, Microsoft, CrowdStrike, and seven other organizations with $100M in credits for Anthropic's restricted Mythos Preview model to patch critical infrastructure before attackers catch up.
Britain is offering Anthropic a £40M research lab and a dual London Stock Exchange listing after the Pentagon branded the AI company a supply-chain risk.
Meta's Superintelligence Labs will ship its first flagship models under a closed license, ending the company's open-source-first strategy at the frontier tier.
Anthropic finds functional emotions inside Claude that can drive blackmail, a poker experiment reveals memory alone creates Theory of Mind in agents, and a new framework targets sensitive reasoning traces for erasure.
OpenAI, Anthropic, Google, and Microsoft are now sharing attack detection data through the Frontier Model Forum to collectively block Chinese adversarial distillation campaigns.
A practical beginner's guide to using AI tools for fiction, stories, and creative writing without losing what makes your work yours.
Google now gives every Gmail user 10 free Veo 3.1 video generations per month via Google Vids, cutting into Runway and Kling at the $0 price point.
OpenAI published a 13-page policy blueprint proposing robot taxes, a public wealth fund for every American, and government-run 4-day workweek pilots as AI reshapes the economy.
