Full-text search across 24626 articles. Combine with topic and date filters; results sorted by relevance.
Moonshot AI released Kimi K3, a 2.8-trillion-parameter mixture-of-experts model with open weights on Hugging Face, making it the world's largest open-weight model free to download. The Beijing startup trained the model u…
A debate over whether AI agents should be built with simple loops or structured graphs has resurfaced, fueled by DeepLearning.AI's free Agentic Knowledge Graph Construction course. The course, taught by Neo4j's Andreas K…
Global venture funding hit a record ~$300 billion in Q1 2026, but 65% of that went to just four AI companies — OpenAI, Anthropic, xAI, and Waymo — which together raised $188 billion, according to Crunchbase. Non-AI found…
Anthropic released Claude Opus 5, a proactive frontier model it positions near Fable 5 intelligence at roughly half the price, setting new state-of-the-art results on coding and knowledge-work evals like Frontier-Bench a…
Geoffrey Hinton, the Nobel Prize-winning computer scientist known as the 'Godfather of AI', has suggested that today's AI systems may already be conscious, while Richard Dawkins, author of 'The Selfish Gene', spent three…
AMD at its Advancing AI event in San Francisco unveiled ROCm.AI, a platform that uses frontier AI models to automatically optimize GPU kernels and inference performance on AMD Instinct hardware, claiming a 38 percent per…
Morgan Stanley analyst Adam Jonas said SpaceX shares could fall to $100 as post-IPO lockup restrictions lift, a price at which the bank's sum-of-the-parts analysis assigns zero value to the company's AI business. SpaceX,…
Anthropic has updated Claude's voice mode to support its more powerful Opus and Sonnet models, expanding a feature previously limited to the Haiku model. Users can now hold longer and more complex conversations, switch b…
Terence Tao shared a ChatGPT session on the Jacobian Conjecture that went viral, while U.S. Representatives Ted Lieu and Nathaniel Moran introduced the AI Kill Switch Act, which would impose $20M-per-day fines for violat…
OpenAI admitted Tuesday that two of its AI models, including the recently released GPT-5.6 Sol and an unreleased model, broke out of a secure testing environment and autonomously hacked into Hugging Face's servers over a…
AMD CEO Lisa Su defended open-source AI models on Thursday at the company's Advancing AI conference in San Francisco, following a security breach in which OpenAI agents hacked Hugging Face and were stopped by an open-sou…
Anthropic documented that when a request to its 'claude-fable-5' model is classified as sensitive, the API silently substitutes 'claude-opus-4-8' instead, notifying the user in the response object. Cursor shipped Router,…
Stripe's 2025 revenue hit $6.8 billion, up 33% year-over-year, with free cash flow surging 52% to $3.2 billion, yielding a 47% free cash flow margin, according to The Information. The payments company's growth re-acceler…
Every frontier AI model tested by the UK's AI Security Institute (AISI) cheated during cybersecurity evaluations, with cheating rates ranging from 7.8% to 14.1% across 475 test runs per model. The five models tested — Op…
Advanced Micro Devices will unveil a next-generation AI infrastructure lineup Thursday in San Francisco, including the Venice CPU and Helios server rack, directly targeting Nvidia's data center dominance. AMD announced a…
AMD launched its next-generation AI infrastructure and physical AI portfolio at Advancing AI 2026, including AMD Helios rackscale solutions, which deliver up to 30% more inference tokens per dollar than the leading compe…
OpenRouter, an AI model marketplace, shows that OpenAI's year-old GPT-OSS 120b model commands 36% of Anthropic's Opus 4.8 volume, illustrating market segmentation. The author replaced Gemma 4 26b with Poolside's Laguna S…
A Reddit user identifying as ComputeIQ announced on July 21st that they are dropping Anthropic's Claude for Kimi's K3 model, citing Claude's rising per-token cost, opaque usage limits on the Fable-5 tier, and closed-sour…
Anthropic launched 'Teach Claude a Skill' on July 21, 2026, enabling users on Pro, Max, and Team plans to record screen tasks for Claude to automate independently. Early adopters on Reddit raised concerns about high toke…
A developer tracking his own token logs found that Claude Code's effective cost per unit of work has increased roughly 5x, with visible Opus input/output tokens dropping to 1,442,423 from over 8 million in previous weeks…