44 Minutes, 176k Tokens: Fable 5.1 Cracks a 370-Year-Old Cipher
Vals AI gave Claude Fable 5.1 Urquhart's unsolved Cyphral Distich: solved in 44 minutes, 176k tokens, zero interjections — and it went on to decode the 285-number Octastich too.
#industry-signal
19 posts
TPU 8i Adds 288GB HBM and 384MB SRAM for Agent Inference
Google's eighth-gen TPUs split into TPU 8t for training (121 ExaFlops, 9,600 chips) and TPU 8i for inference (288GB HBM, 80% better perf-per-dollar).
Atlassian Align Tool Dropped — Vendor Clears Jira Cloud
Atlassian retires Align on 2027-03-15, pushing roadmap work into Jira Cloud. Migration window open until then.
10,000 Agents in 88 Hours Prove a 90-Year-Old Navier-Stokes Problem
An unnamed internal model plus ~10,000 coordinating agents closed the Navier-Stokes Millennium Problem in ~88h (130B tokens), with a verified Lean proof. The capability step OpenAI's flagging.
GPT-6 Astra Shifts the Frontier Model Calculus—99.9% on ARC, 100% on Exploits
GPT-6 Astra saturates benchmarks and raises the bar on computer use, coding, and cybersecurity. Teams must now reconsider whether open-weights or smaller fine-tuned agents remain cost-effective.
NVIDIA's Custom HBM Signals Memory Bandwidth Is the New Constraint
NVIDIA's NVHBM moves memory control to the HBM stack itself, not the GPU. This architectural shift reveals memory bandwidth, not compute cores, is the real scaling bottleneck for AI inference.
Samsung zHBM Cuts I/O Power ~70% — Memory and Compute Merge by 2028
Samsung's three-phase HBM roadmap moves the memory controller into the base die, then stacks the whole DRAM on the processor — cutting I/O power ~70% at zHBM.
Opus Is Pro+-Only Now That Copilot Pro Meters Every Call
Copilot Pro now meters every model call, Opus moved to Pro+ only, and limits split 5X between tiers — budget for metering.
Meta's Coding Agent Is Cheap Because Your Data Funds It
Meta's Muse Code beta ships with worktree isolation and a crash-safe event log — but its $0.10/M contributor tier runs on your training data.
Your Nova Deployments Are Going Away — Plan the Migration Now
Amazon is deprecating Nova Premier, Omni, Reel and Canvas in a strategic pivot. Audit your Bedrock calls now, before a deadline lands.
16,000 MT/s DDR5 MRDIMMs Just Raised the AI Memory Ceiling
Renesas' Gen 3 MRDIMM pushes server DDR5 to 16,000 MT/s — 25% over Gen 2 on existing boards. Memory-bound decode gains a real ceiling.
Your AI Stack Choices Lock You Into Years of Vendor Control
Sovereign AI isn't data residency. It's territorial, operational, technological, legal—and financial control. The three case studies that changed CFO thinking in Q1 2026.
The End of On-Demand Frontier AI
Google's $920M/month SpaceX contract shifts AI infrastructure from on-demand cloud to locked-in capacity. Here's what changes for enterprise architecture.
NVIDIA Halves Vera Rubin Memory — LPDDR5X Shortage Hits First
NVIDIA's next-gen Vera Rubin drops standard CPU memory from 54TB to 28TB per rack. Here's what the LPDDR5X shortage means for AI infrastructure.
Your API Costs Are About to Correct
OpenAI's IPO filing ends the era of subsidized frontier AI. Here's what the pricing shift means for local inference, enterprise contracts, and your agentic stack.
NVIDIA's First Server CPU Clears the Performance Floor
NVIDIA's Vera CPU benchmarks show it competes directly with EPYC and Xeon, but software maturity and platform lock-in dictate adoption.
DDR5 Tripled Since October — Buy Nothing Yet
DRAM and NAND prices have surged 90–130% since late 2025. A Samsung factory walkout and HBM reallocation mean the crunch won't clear before late 2027 at earliest.
Claude Code v2.1.100+ Burns ~20K Phantom Tokens Per Request
A server-side bug in Claude Code v2.1.100+ inflates every request by roughly 20K cache_creation tokens — about 40% overhead. Pin v2.1.98 until fixed.
Third-Party Claude Agents Lose the Subscription Subsidy June 15
Anthropic is splitting Claude billing on June 15 — Agent SDK and ACP usage moves to a capped credit pool ($20/$100/$200) at full API rates.