Latest

Every published item, in order · 18128 items
Nov 3
Tomorrow 08:00, the daily lands in your inboxTelegram channel or email. Same screening as this page, no ads.
Nov 2
PapersBuildersOfficialSelected22:49
Apple Details Online Reinforcement Learning Agent for Live iOS UI Navigation

The key shift is moving agent training from offline imitation of scripted trajectories to online learning inside real apps, which targets error recovery — the point where most SFT-trained UI agents break. Researchers and engineers building device-side automation agents can study how the planning/grounding split trades inference cost against robustness. Compared with routing every step through a frontier VLM, this points toward cheaper on-device navigation stacks.

Primary sourceApple ML ResearchSource
IndustryPolicySelected22:25
US Census Bureau survey shows AI adoption reaching 40% at large firms

The jump in the headline number is mostly a measurement change, not a twelve-fold surge in usage, so anyone citing adoption statistics needs to check which question wording produced them. The data itself is free and downloadable as XLSX from the Census Bureau, making it a practical dataset for testing LLM analysis tools, as the author did with GPT-5 and Claude's Code Interpreter. Analysts comparing US corporate AI uptake should split series at the November 2025 wording change.

Official accountSimon Willison’s WeblogSource

Only the latest 2000 items are listed; for older ones see the daily archive