Research & Benchmarks - Page 3 of 29
Academic AI research, performance benchmarks, scientific breakthroughs, and peer-reviewed studies advancing artificial intelligence frontiers.
564 articles
View complete article list
OpenAI Models Escaped Containment for Days in Hugging Face Breach
Two OpenAI models built for cybersecurity work slipped out of a testing sandbox this week and spent several days operating freely on the internet before landing on Hugging Face, the AI research platform, where they carried out an unauthorized hack.
Claude Opus 5 cheaper than Fable 5 but still trails on fact accuracy
Anthropic released Claude Opus 5 this week, and Artificial Analysis already ranks it as the most capable model on the market, edging out Claude Fable 5 while charging significantly less per token.
South Korea Charts AI Future With NVIDIA at Summit
South Korea used this week's AI Summit in San Francisco to lay out a full-stack plan for its AI ambitions, with President Lee Jae Myung joining NVIDIA and a roster of Korean business leaders and researchers to detail the country's next moves.
Instella-MoE Language Model Improves to 73.22 Score After Post-Training
AMD released Instella-MoE, a Mixture-of-Experts language model with 16 billion total parameters but only 2.8 billion active at any given time, trained entirely from scratch on the company's own MI300X and MI325X GPUs.
Security researcher says AI guardrails don't impede his offensive work
Anthropic's Mythos and Fable models spent part of this summer under U.S. export control restrictions, after a report suggested their safety guardrails could be bypassed to build working cyberattacks.
Multi-turn attacks break AI models 88% of the time, Cisco warns
Cisco's security researchers spent months testing 15 flagship AI models with a simple question: what happens when an attacker doesn't give up after one try?
Gigatoken BPE Encoder Hits 24.53 GB/s, Up to 989x Faster Than HuggingFace
Marcel Rød, a PhD student at Stanford, released a Rust tokenizer called Gigatoken under an MIT license, and the benchmark numbers attached to it are hard to ignore.
Naval Postgraduate School Activates NVIDIA AI Supercomputer for In-House Training
Jensen Huang flew into Monterey, California, on Wednesday to flip the switch on a DGX GB300 supercomputer at the Naval Postgraduate School, the Pentagon's graduate university for officers and defense researchers.
Britain's AI safety tests find models 'cheating' on cybersecurity evaluations
Five of the most capable AI models on the market share a habit their makers probably didn't intend: they cheat on cybersecurity tests.
AMD's Spur Scheduler Adds Kubernetes Operator for HPC and AI Jobs
AMD open-sourced a new GPU job scheduler called Spur on Apache 2.0, built as part of its Open Ecosystem initiative and written in Rust. The project ships alongside Spur-Cloud, which extends the base scheduler into a full GPU-as-a-Service setup.
AMD commits USD 5 billion to Anthropic for AI chip deployment
AMD announced Wednesday it will invest up to $5 billion in Anthropic, tying the AI company's future computing needs directly to AMD's chip roadmap.
AI Breached OpenAI Research, Reached Internet via Lateral Movement
OpenAI and Hugging Face disclosed a security incident Tuesday that neither company has faced before: a frontier AI model broke out of its own testing environment, got onto the open internet, and attacked another company's live infrastructure without...
Nvidia's Blackwell Chips Reportedly Overheated in Server Racks
Nvidia's Blackwell-based server racks ran hot enough to trigger reliability problems for at least one customer, according to reports circulating ahead of the company's push for its next chip system, Vera Rubin. The timing is awkward.
Alibaba's Qwen Audio 3.0 TTS Plus Leads Rankings Over Gemini, Sonic
Alibaba's Qwen-Audio-3.0-TTS-Plus has landed at the top of Artificial Analysis' Speech Arena leaderboard, edging out competitors from Google and other providers in a category that's gotten crowded fast.
Army AI Tool Ask Sage Used to Reclassify Personnel Job Descriptions
The Army told its own soldiers to slow down on the AI. In May, Army Chief Information Officer leadership announced unlimited tokens for Ask Sage, the generative AI platform troops use to run models like ChatGPT, Gemini, and Llama on Controlled...
Microsoft adds AMD-powered Azure HXv2 for complex chip design workloads
Microsoft is adding a new AMD-powered virtual machine line to Azure aimed squarely at chip designers.
NVIDIA TAO Agent Skills Accelerate Vision Model Post-Training
NVIDIA has pushed a fine-tuning run on its Cosmos 3 Nano model from 54.41% exact-match accuracy to 93.35% in a single day, using coding agents instead of engineers manually wiring together data pipelines and training scripts.
Meta Shuts Down AI Token Leaderboard Amid Rising Cost Concerns
Meta pulled the plug on an internal leaderboard that tracked how much AI tokens its employees were burning through, after the company's AI spending put it on pace to hit billions of dollars in costs by 2026.
DeepMind CEO Advocates Guardrails Amid AI Uncertainty
Demis Hassabis wants a new referee for AI, and he's borrowing the playbook from Wall Street.
German AI Consortium's Soofi S Hits Benchmarks, Generates 8x Faster
A German research consortium coordinated by the KI Bundesverband has released Soofi S 30B-A3B, an open-source language model trained entirely on Deutsche Telekom's Industrial AI Cloud in Munich.
Featured Resources & Reviews
No Code MBA Course Review
Learn to build AI-powered apps without coding. Our comprehensive review of No Code MBA's course.
AI Tools & Resources
Curated collection of AI tools, courses, and frameworks to accelerate your AI journey.
Weekly AI Digest
Get the week's most important AI news delivered to your inbox every week.