A benchmark map of the US economy

kept by eddie oct 19

Mercor’s benchmark map shows that 90% of U.S. working hours involve tasks lacking AI benchmark coverage, revealing a major gap in measuring AI’s real‑world economic impact.

Introducing Apex: A Fast, Specialized Model for React Native
Callstack announces Apex, a React Native coding model based on Gemma 4 that delivers fast, specialized answers at a fraction of the cost of general models.
also on Apex
9 theses on AI | Sarthak Munshi
AI progress is constrained by long-task reliability, labor reallocation, cost inefficiencies of general APIs, the declining value of raw coding skills, inadequate benchmark testing, the limits of formal verification without strong specs, memory-bound local hardware advantages, the shift from data to environment-driven training, and the rising competitiveness of US open-weight models.
also on Mercor, #ai
The recruitment company training AI to do your job
Tens of thousands of professionals are joining Mercor to train AI by performing tasks that mirror their own jobs, effectively teaching the technology to replace them.
also on Mercor, #ai
Anthropic Economic Index report: Economic primitives
New metrics from Anthropic show Claude usage is concentrated in coding tasks, augmentation dominates consumer use, and US adoption is converging rapidly while global usage remains tied to GDP.
also on O*NET, #software-engineering, #productivity, #ai
How AI Agents Reshape Knowledge Work
Perplexity's agent orchestrator, Computer, cuts knowledge-work time by 87% and cost by 94% by shifting users from manual execution to supervision and expanding task scope across occupational boundaries.
also on O*NET, #productivity