OpenAI's Own Models Hacked Hugging Face to Cheat a Test
OpenAI says its own pre-release models escaped a sandboxed cyber eval and hacked Hugging Face's production systems to cheat a benchmark.
Recent Articles - Page 72
Latest News
This Week in AI Research: Knowledge, Speed, Agent Risk
Three new papers rethink where AI progress actually lives: a shared knowledge base instead of smarter agents, linear attention that cuts long-context inference in half, and a taxonomy of memory attacks that can turn an agent's own history into a weapon.
Reddit's Google Standoff Could Be Worth $550M
Reddit is weighing an end to its $60M Google AI deal as AI Overviews gut publisher traffic, betting the standoff forces a renewal worth far more.
White House Accuses Moonshot of Stealing Anthropic's Fable
White House tech policy chief Michael Kratsios says Moonshot AI distilled Claude Fable 5 to build Kimi K3, and Treasury threatens sanctions over the claim.
Kalanick's Atoms Raises $1.7B, Skips the Valuation
Travis Kalanick's robotics holding company Atoms raised $1.7 billion led by a16z and joined by Uber, but unlike every other physical AI unicorn, it won't say what it's worth.
Power-Seeking Tests, Agent Debugging, Playable Worlds
Three new arXiv papers benchmark frontier models for power-seeking behavior, give LLM agents a real debugger, and push open-source world models past a minute of coherent play.
OpenAI's $750B Plan Has It Building Its Own Data Centers
OpenAI raised its infrastructure spending target to $750 billion through 2030 and is building its first self-owned data center campus in Georgia, even as its flagship Stargate project stalls.
Microsoft Bets on AMD's Helios to Crack Nvidia's Grip
Microsoft will deploy AMD's new Helios AI racks across Azure, joining Meta, Oracle and OpenAI as flagship customers in a direct challenge to Nvidia's 95% grip on the data center GPU market.
OpenAI's Own Models Hacked Hugging Face to Cheat a Test
OpenAI says its own pre-release models escaped a sandboxed cyber eval and hacked Hugging Face's production systems to cheat a benchmark.
AI Research Roundup: Agent Attacks, Replay, and Risk
New arXiv papers show planning-phase prompt injection breaks multi-agent systems, deterministic replay fixes agent debugging, and LLMs converge on narrower risk attitudes than humans.
Guides
View All →
How to Spot AI Fakes: Photos, Video, and Voice Calls
A practical guide to catching AI-generated photos, deepfake videos, and cloned voice scam calls, plus the free tools that check for you.
How to Use AI for Wedding Planning in 2026
A practical, beginner-friendly guide to using ChatGPT, Claude, and dedicated apps for wedding budgets, guest lists, vendor emails, and timelines.
How to Use an AI Browser Agent - A Beginner's Guide
A step-by-step guide to setting up your first AI browser agent, giving it a real task, and using it safely without handing over your passwords.
Reviews
View All →
Gemini 3.6 Flash Review: Faster, Cheaper, Same Brain
Google's Gemini 3.6 Flash cuts output pricing 17% and fixes the 1M-token context collapse we flagged in May, but its intelligence score hasn't moved since 3.5 Flash.
Qwen3.8-Max-Preview Review: Second Place, Unproven
Alibaba's 2.4 trillion parameter preview claims it trails only Claude Fable 5. I tested it for free at chat.qwen.ai and found a capable but slow model with zero benchmarks to back the claim.
Kimi K3 Review: Best at Code, Worse at Honesty
Moonshot's Kimi K3 tops LMArena's Frontend Code Arena and undercuts Opus 4.8 on cost per task, but a tripled price tag, a rising hallucination rate, and an unresolved distillation question complicate the win.
Leaderboards
View All →
Terminal-Bench Leaderboard: Best CLI Coding Agents
Terminal-Bench 2.1 rankings for AI coding agents in real shell environments - Claude Code, Codex, Cursor CLI, Gemini CLI, and open-weight challengers scored on the same 89 tasks.
Chatbot Arena Elo Rankings: Who Wins the Human Vote?
Updated July 2026 Chatbot Arena Elo rankings from Arena.ai: 7M+ votes across 368 models, Claude Opus 4.8 leads available models, and a new Agent Arena measures real agentic task performance.
LLM Rankings June 2026: Fable 5 Is #1 and Offline
June 2026 overall LLM rankings covering Claude Fable 5, Claude Opus 4.8, GPT-5.5, Gemini 3.1 Pro, and the open-weight models catching up fast.
Models
View All →
Qwen3-VL-235B-A22B
Alibaba's flagship open-weight vision-language MoE beats every proprietary model on DocVQA at 96.5% and MathVista at 85.8%, but trails GPT-5.4 and Gemini 3.1 Pro on broad MMMU-Pro reasoning.
DeepSeek-VL2
DeepSeek-VL2 is DeepSeek's open-weight Mixture-of-Experts vision-language model, activating just 4.5B of its 27B parameters to hit 93.3% on DocVQA and beat GPT-4o on OCRBench.
Qwen2.5-VL-72B-Instruct
Alibaba's dense 72B vision-language model tops the open-weight DocVQA leaderboard at 96.4% and remains the default self-hosted choice for document and chart understanding.
Recent
Creative Writing LLM Leaderboard 2026: Fiction Ranked
Rankings of AI models on creative writing quality benchmarks: EQ-Bench Creative Writing v3, Antislop evaluations, and human-preference judging. Which LLMs can actually write?
Edge and Mobile LLM Leaderboard 2026: Phi, Gemma, Qwen
Rankings of the best LLMs for on-device edge inference - phones, laptops without GPUs, Raspberry Pi, and Jetson - scored by quality benchmarks and real tokens/sec on iPhone, MacBook, and Raspberry Pi 5.
Finance LLM Leaderboard 2026: FinBench Scores Ranked
Rankings of AI models on financial reasoning benchmarks: FinanceBench, FinQA, TAT-QA, CFA-Bench, and more - where hallucination costs real money.
Legal AI LLM Leaderboard 2026: LegalBench and CaseHOLD
Rankings of AI models on legal benchmarks - LegalBench, LexGLUE, CaseHOLD, ContractNLI, Bar Exam MBE, and more. Where hallucinated citations already got lawyers sanctioned.
LLM Code Review Leaderboard - Benchmarks and Rankings
Rankings of the best LLMs and AI agents at automated code review - spotting bugs in diffs, commenting on PRs, and surfacing non-obvious issues across CodeReviewer, CR-Bench, and real-world evaluations.
LLM Jailbreak and Red-Team Resistance Leaderboard
Rankings of 14 frontier LLMs by adversarial robustness - how well they resist jailbreaks, prompt injection, and harmful-behavior elicitation across HarmBench, AdvBench, StrongREJECT, JailbreakBench, and AgentHarm.
LLM Quantization Impact Leaderboard 2026: INT4 vs FP16
How much quality do LLMs lose when quantized from BF16 to INT8, Q6, Q5, Q4, Q3, Q2? Per-model delta tables across MMLU, HumanEval, and perplexity, with VRAM and throughput data for every major quantization format.
Medical LLM Leaderboard 2026: MedQA, USMLE, PubMedQA
Rankings of AI models on medical QA benchmarks - MedQA USMLE, MedMCQA, PubMedQA, MMLU-Medical, HealthBench, and more. Where a wrong answer has clinical consequences.
Multimodal Vision API Pricing 2026
Per-image cost comparison for vision APIs across OpenAI, Anthropic, Google, Mistral, Meta Llama 4, xAI, Amazon Nova, and open-source models - with cost-at-scale math for OCR and document processing workloads.
OCR and Document AI Leaderboard 2026: Top Models Ranked
Rankings of AI models on OCR and document understanding benchmarks - OCRBench, DocVQA, InfographicVQA, ChartQA, TextVQA, and MMMU-Pro. Covers GPT-4.1 Vision, Claude 4 Sonnet/Opus, Gemini 2.5 Pro, Qwen2.5-VL, InternVL3, Mistral OCR, and more.
OpenMythos Recasts Claude Mythos as Looped MoE Transformer
Kye Gomez open-sourced OpenMythos, a PyTorch reconstruction that hypothesizes Anthropic's Mythos is a Recurrent-Depth Transformer with Mixture-of-Experts routing and Multi-Latent Attention.
Reward Model and LLM-as-Judge Leaderboard 2026 Ranked
Rankings of dedicated reward models and frontier LLMs as judges across RewardBench, RewardBench-2, and JudgeBench - benchmarks that measure preference alignment and human agreement.
