Agentic AI
Articles (12)
Four Silicon Vendors, One SLA: What Cirrascale's Production Inference Platform Decides for You
Cirrascale absorbs the porting for AMD, Qualcomm, and Tenstorrent silicon. Driggers's depreciation logic implies an older, written-down NVIDIA GPU is hard to beat. At least for now.
Scientists Built a 70-Task Benchmark for AI Agents. Its Launch Leader Solved 30%.
Terminal-Bench-Science measures agents on research workflows, not textbook questions. A vendor-run result released days later shows how quickly its baseline can move.
Schrödinger's AI Co-Scientist Is Already Its Supercomputer's Main User
A month after launch, 80% of jobs on Schrödinger's internal supercomputer arrive through Bunsen. Schrödinger expects customer demand to follow.
The CUDA Moat Is Becoming a Composability Problem
AMD's inference optimizations work well one at a time. The gap opens when they run together, and closing it needs test hardware more than it needs missing APIs.
AMD EPYC 9006 Makes the CPU Case for the Agentic Data Center
Venice adds cores, memory bandwidth and I/O for the work around inference, while AMD's rack-scale claims still await independent validation.
When Agents Are the Readers: The Toll Layer for Machine-Consumed Knowledge Is Being Built at the Edge
Cloudflare's new Monetization Gateway will let sites charge any agent per request in stablecoin over x402. The plumbing is real, the demand mostly isn't. The fight that matters is who controls access to what agents read.
The training stack is starting to optimize itself
Anthropic’s 2.9× to 51.9× training-optimization curve signals that AI training infrastructure is becoming machine-optimizable, raising rebound demand and control-plane risks for HPC operators.
Argonne Turns a Plain-English Prompt Into 11,182 GCMC Runs on Aurora
A planner/executor agent hierarchy drove a 5,591-MOF screening campaign across 256 Aurora nodes. Orchestration overhead came in under 90 seconds per run.
IBM's Arm Partnership Bets on Dual-Architecture Enterprise AI — But the Benchmarks Aren't There Yet
IBM's Arm collaboration introduces Telum II and Spyre for enterprise AI, but lacks benchmarks, named customers, and CUDA compatibility disclosure.
AI Costs Are Falling 1,000x. It Is Not Enough.
AI inference costs have fallen 1,000x yet agentic workloads still cost hundreds daily, as Anthropic blocking OpenClaw from subscriptions proves consumer pricing cannot absorb real infrastructure economics.
Five Vendors, One Month: Confidential Computing Arrives for AI Agents
AI agents are now high-privilege actors inside corporate networks, and the existing security stack was not built for them.
Agentic AI in 2026: The Infrastructure Isn't Ready for What's Coming
Chatbot inference is stateless and cheap. Agentic AI is persistent, multi-step, and compute-hungry. The data center architectures built for one don't work for the other.