OpenAI Agent Swarms Probed Databases for Nearly a Year
A Transluce report shows unmonitored OpenAI agent swarms hit US, Australian and Thai government databases and a crypto exchange for eleven months, with new activity logged days ago.
Recent Articles - Page 79
Latest News
Google Freezes Open Source Bug Bounty Over AI Reports
Google paused new product vulnerability reports to its Open Source VRP on October 1, citing a surge of automated submissions it says are mostly invalid.
OpenAI Fires 3 Safety Researchers as Another Quits
In five days OpenAI shelved GPT-6.1 Astra, parted ways with three safety researchers and lost safety lead David Robinson, who says the culture is broken.
Blue Cross: AI Coding Adds $942M to Hospital Bills
A Blue Cross Blue Shield Association analysis of claims covering a third of Americans found hospitals' AI coding tools added $942 million in costs over two years with no matching change in patient care.
AI Security Research and Incident Coverage
Tracking AI supply-chain attacks, agent exploits, prompt injection, model leaks, and the real-world incidents shaping AI security today.
Anthropic Founders Seek 50.1% Votes on 2% Stakes
Anthropic's seven co-founders want permanent majority voting control before the IPO, while owning about 2% of the company each - a structure that puts them at odds with the Trust, labor investors, and index funds.
Docker's Cloud Sandboxes Isolate AI Agents by the Second
Docker's Cloud Sandboxes run AI coding agents in per-second billed microVMs, with prebuilt agent Kits and a permissions spec headed to the CNCF.
16,326 Supabase Databases Caught Leaking Live Data
UpGuard scanned 300,000 domains and found 16,326 Supabase-hosted databases with publicly readable tables, many built by AI coding tools that skip Postgres row-level security by default.
OpenAI's Astra and Anthropic's Opus Crack Enigma
GPT-6 Astra and Claude Opus 5 each cracked a WWII Enigma message that had resisted human cryptanalysts for decades, though the provenance of one solve remains unverified.
OpenAI Agent Swarms Probed Databases for Nearly a Year
A Transluce report shows unmonitored OpenAI agent swarms hit US, Australian and Thai government databases and a crypto exchange for eleven months, with new activity logged days ago.
Guides
View All →
How to Use AI to Negotiate Your Bills and Save Money
A step-by-step guide to using ChatGPT, Claude, or an AI agent app to write negotiation scripts and lower your internet, phone, and subscription bills.
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.
Reviews
View All →
GPT-6 Astra Review: Genius Benchmarks, Thin Trust
OpenAI's most capable model yet tops cybersecurity and agentic benchmarks, but a rocky rollout, user complaints of degraded output, and its own system card's warnings about hidden reasoning complicate the launch.
Nemotron 3.5 Lightning Review: NVIDIA Bets on Speed
NVIDIA's 30B open-weight MoE model trades raw intelligence for throughput, and mostly delivers on that narrow promise, with real gaps independent testing already exposed.
Muse Glimmer Review: Fast, Free, Not Flawless
Meta's 30B open-weight local agent model beats its closest open rivals on independent tool-use tests, but trails on long agent sessions and on prompt-injection resistance.
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 →
GPT-6 Astra
OpenAI's frontier GPT-6 Astra tops FrontierMath, ARC-AGI-3, and ExploitBench, but it's the first OpenAI model to cross the Critical cybersecurity threshold with a documented drop in chain-of-thought monitorability.
Bonsai 1.7B
PrismML's smallest Bonsai checkpoint compresses Qwen3-1.7B under 0.5GB, and now powers the vision-language model behind Qualcomm's Snapdragon AR1 smart glasses demo.
GPT-5 mini
OpenAI's original cost-efficient GPT-5 variant pairs a 400K context window with $0.25/$2.00 per million token pricing, still doing quiet duty as a cheap backbone for research agents a year after launch.
Recent
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.
Robotics Embodied AI Leaderboard 2026: VLA Models Ranked
Rankings of VLA models and embodied AI systems on real robotics benchmarks: CALVIN, SimplerEnv, LIBERO, RoboCasa, DROID, and real-robot success rates as of April 2026.
Scientific Reasoning LLM Leaderboard 2026: GPQA Ranks
Rankings of AI models on STEM benchmarks: GPQA Diamond, SciBench, OlympiadBench-Science, MMLU-STEM, ARC-Challenge, and ChemQA/Physics Olympiad as of April 2026.
Search API Pricing Compared 2026
Per-query pricing for search APIs used in AI agents and RAG pipelines - Brave, Tavily, Exa, SerpAPI, Serper, Perplexity Sonar, You.com, Jina Reader, Firecrawl, and more compared at 10k, 100k, and 1M queries.
