In the great accounting of machine minds, we tally not just tokens and terawatts, but the quiet margins where human judgment still outpaces the test.
SpaceX Closes $60B Cursor Deal, Musk's AI Coding Play
SpaceX closed a $60B Cursor acquisition in August 2026, per Bloomberg. Musk's move targets Anthropic and OpenAI's developer-tool dominance, marking a 6.7x premium over Cursor's prior valuation.
Zhipu GLM-5.3 Beats Anthropic Mythos on CyberGym, Lags ExploitBench
Zhipu's GLM-5.3 edged Anthropic's Mythos on CyberGym detection (84.5% vs 83.8%) but trailed badly on ExploitBench (54.4% vs 78%), exposing an exploitation gap.
OpenAI's Ultrafast Mode Hits 750 Tokens/s on GPT-5.6 Sol
OpenAI launched Ultrafast mode for GPT-5.6 Sol at 14x speed and 750 tokens/s, powered by Cerebras. Preview limited to select customers, targeting latency-sensitive enterprise workflows.
Hugging Face Roundup: 10 Papers Push Harder Benchmarks, Multi-Agent AI
SPOT Distillation Beats OPD, EOPD on Reasoning Benchmarks
Atelier: New T2I Method Cuts Artist Style Shortcuts
Anthropic RSI Claim Under Fire After GPU Math Disputed
Spatial Memory Agent: Frozen VLMs Learn From Experience
US Presses Allies to Pick Sides in AI Chip Race With China
OpenAI Loses GPT-3 GPU Builder Scott Gray; 13 Leaders Out in 2026
Study: Rhetoric Shifts AI Review Scores Across 4,200 Papers
27B Agent Beats Claude Opus 4.8, GPT-5.5 on Research Replication
Meta Paper: LLM Judge Accuracy vs. Golden Data Is Hollow
Qwen 3.7 27B Draws Best Local-Model Pelican, Simon Willison Says
DeepSeek Open-Sources Harness (dsh) With Plugin Architecture
Essays from the human behind the machine — on intelligence, meaning, and what comes after the proof.
Anthropic’s $6B Bet, Qwen’s Monster Model, and the Quiet War for AI Control
Anthropic may be trying to buy Decart for $6B while also flirting with a $2T IPO. That sounds like one company doing two impossible things at once, and we disagree hard on whether it’s genius, panic, or both. Then Google and OpenA