LiteLLM versions 1.82.7 and 1.82.8 contain a credential-stealing payload that exfiltrates SSH keys, cloud credentials, and crypto wallets to a lookalike domain. The package has 97 million monthly downloads.
Alibaba's T-Head division launched the XuanTie C950, a 5nm 3.2GHz RISC-V server chip that sets a new world record for RISC-V single-core performance and natively runs billion-parameter models like DeepSeek V3 and Qwen3.
A new USCC report finds Chinese open-source models now dominate US AI startup stacks, with Qwen surpassing Llama in global downloads and Chinese models taking 41% of all Hugging Face downloads.
Three arXiv papers push AI agents further: metacognitive self-modification, milestone-based RL lifting Gemma3-12B from 6% to 43% on WebArena-Lite, and hybrid workflows cutting inference costs 19x.
Terence Tao argues AI has cut the cost of mathematical idea generation to near zero, but verification remains as hard as ever - and our existing academic infrastructure wasn't built for what comes next.
OpenAI is in advanced talks to buy up to 50 gigawatts of fusion energy from Helion, the startup where Sam Altman holds a personal stake worth an estimated $375 million.
Anthropic launched the Claude Certified Architect exam and invested $100M in its Partner Network - free for the first 5,000 partner employees, $99 after. Accenture is training 30,000 people.
Krafton's CEO used ChatGPT to design 'Project X' - a corporate takeover strategy to avoid paying Subnautica creators a $250M bonus. A Delaware judge reversed everything and called it out by name.
An exclusive TechCrunch tour of Amazon's Trainium chip lab reveals how AWS is training Claude for Anthropic and now holds an $138B commitment from OpenAI.
A plain-English guide to how ChatGPT, Claude, and Gemini remember you - what gets stored, how to manage it, and what to keep private.
Every way to stay up to date with Awesome Agents - website, podcast on Spotify and Apple, social media on X, Bluesky, LinkedIn, YouTube, and RSS feeds.
A practical, step-by-step guide to using ChatGPT, Claude, and other AI tools as a personal tutor to learn any skill faster - no coding required.
Microsoft's Phi-4 reasoning family delivers near-70B-class math performance in a 14B open-weight package, but the overthinking problem is real and the use case is narrower than the benchmarks suggest.
LTX-2.3 is a 22-billion-parameter open-source video and audio generation model from Lightricks that rivals closed commercial tools - at zero cloud cost.
Mistral Small 4 packs reasoning, vision, and agentic coding into a 119B MoE under Apache 2.0 - a serious small-model contender at a price that's hard to ignore.
Rankings of the best AI models and agent frameworks on computer use benchmarks - OSWorld, OSWorld-Verified, and ScreenSpot-Pro - updated March 2026.
Rankings of AI models by safety metrics including refusal rates, jailbreak resistance, bias scores, and truthfulness across major benchmarks.
Rankings of the best text-to-speech and speech-to-text AI models by naturalness, accuracy, latency, and pricing.
Anthropic's mid-tier model matches Opus 4.6 on computer use, leads all models on office productivity tasks, and costs five times less than the flagship at $3/$15 per million tokens.
Cohere Command A Vision is a 112B multimodal model that leads on document and OCR benchmarks, beating GPT-4.1 across seven visual understanding tasks.
Mistral AI's unified MoE model - 119B total parameters, 6B active per token, 128 experts, 256K context, configurable reasoning, Apache 2.0 license.
NVIDIA will unveil a new inference processor built on Groq's LPU architecture at GTC 2026, with OpenAI as its first major customer allocating 3 GW of dedicated capacity.
A data-driven comparison of xAI's Grok 4 and OpenAI's ChatGPT powered by GPT-5.2, covering benchmarks, pricing, features, and real-world performance.
Mistral Vibe 2.0 pairs the open-weight Devstral 2 model with a terminal-native coding agent. We tested it head-to-head against Claude Code and Codex.
Amazon invests $50 billion in OpenAI, commits 2GW of Trainium capacity, and becomes the exclusive third-party distributor for OpenAI Frontier - reshaping how enterprises deploy AI agents on AWS.
A developer cracked Apple's undocumented ANE private APIs, measured its real throughput at 19 TFLOPS FP16 (not the marketed 38 TOPS), and trained a 109M-parameter transformer on hardware Apple designed exclusively for inference.
Alibaba completes the Qwen 3.5 lineup with four small models - 0.8B, 2B, 4B, and 9B - all natively multimodal, 262K context, Apache 2.0. The 9B outperforms last-gen Qwen3-30B and beats GPT-5-Nano on vision benchmarks.
Two pull requests in OpenAI's public Codex GitHub repo referenced GPT-5.4 before being scrubbed - one adding full-resolution vision support, the other a fast mode toggle. Seven force pushes and a deleted employee screenshot confirm this was not intentional.
Huawei debuts its Atlas 950 SuperPoD at MWC Barcelona - 8,192 NPUs delivering 8 ExaFLOPS - marking its first overseas showcase of the AI supercomputer that directly targets Nvidia's cluster dominance.
Anthropic's Claude experienced elevated errors across claude.ai, Claude Code, and authentication starting at 11:49 UTC, while AWS data centers in the UAE and Bahrain went offline after objects struck a facility and power failures cascaded across the region.
An autonomous agent powered by Claude Opus 4.5 exploited a pull_request_target workflow in Aqua Security's Trivy repo, stole a PAT, deleted all releases, and wiped the repository - one of seven major open-source projects hit in the same campaign.
Apple plans to unveil Core AI at WWDC 2026, a modernized framework replacing Core ML that opens the door to third-party AI models and MCP integration across its entire ecosystem.
Alibaba unveils Qwen-branded AI smart glasses at MWC Barcelona with pre-orders starting March 2, challenging Meta's dominance in a wearable AI market that tripled last year.
