Using AI makes your work better. But are you, the person using it, actually getting better? Title: Does better work always mean better workers? URL: Google Research ran a three-month field experiment with 133 patent attorneys to test how AI assistance affects both the quality of work and the durability of professional judgment. Highlight 1 📈 Everyone's work improves with AI Patent drafting with AI access improved quality by 0.34 SD after 10 days and 0.38 SD after 90 days. Junior attorneys also drafted 18 minutes faster than the control group. Highlight 2 ⚖️ Take the AI away, and experience matters At day 90, attorneys reviewed patent claims without AI. Senior attorneys (7+ years) improved by a striking 0.45 SD. Junior attorneys showed no average improvement at all, instead splitting into stronger and weaker performers. Highlight 3 🔍 Fixing vs. rephrasing Junior attorneys worked sequentially from the top, spending time on surface-level wording and rarely fixing the substantive flaws they identified. Senior attorneys used AI as a logic auditor and rebuilt claims from the ground up. Better output and better workers aren't the same thing. #AIWorkforce# #KnowledgeWork#

Knowledge work is SO much harder to automate with agents than code We underestimate how easy software development is for agents - Tons of high quality automated feedback loops (types, tests) - Well-organized free documentation, searchable on the web - Simple-to-use workspaces with version control (i.e. repositories) - Long history of systematising and delegating work (i.e. sprints, tickets, specs) Knowledge work has none of these advantages

An example of useful knowledge work: I assigned GPT-6 to read through tens of thousands of my emails, my writings, my calendar appointments and more to assemble a personal knowledge base of research, contacts, ideas, relationships, and tasks over my recent career. GPT-6 downloaded the appropriate software, figured out a strategy, and built a multi-gigabyte personal wiki without any further intervention from me over the course of five days of uninterrupted work. Twice a day, the AI now goes through my emails, cross-references them with this extensive knowledge base, and sends me a summary of things I should be paying attention to, things I might be interested in, etc. (Two questions you may have: Yes, I gave the AI access to my computer and this involves risks, and you should be careful before you do the same. And I do not take money from any AI lab and pay for my own usage, but during the trial period, I was not charged for tokens, so I cannot tell you how much this process would have cost, but it likely would have been very substantial. The ongoing briefings do not use substantial amounts of token)

Grok 4.7 brings frontier-level knowledge work at a fraction of Fable 5.1’s token cost. You're literally getting 80% lower input pricing and 88% lower output pricing for a minuscule gap in performance. Great work @SpaceXAI and @elonmusk!

Kimi K3.1 Pro is among the leading models in agentic knowledge work and coding, but early signs suggest it will rank lower on AA-3D We are still collecting the preference votes for its Elo on AA-3D, our benchmark for how well AI models build interactive 3D scenes. In agentic knowledge work and coding, it is among the leading models on GDPval-AA and Terminal-Bench 4.0. Its predecessor, Kimi K3, is 12th of 22 models on AA-3D. Here is one of its AA-3D scenes: a 492 m mountain-bike descent with berms, roots, a rock garden, a drop, and a jump. A demo run rides it from top to bottom, and you can take over the steering, pedaling, and braking at any point. Congratulations to @Kimi_Moonshot on the release!

Claude Opus 5.5 (max) is a top performer in agentic knowledge work. It leads AA-Briefcase v1.1 at 1,822 Elo (+143 over Fable 5.1) and GDPval-AA v2.1 at 1,846 Elo (+111 over Claude Fable 5.1, +138 over Claude Opus 5). On AA-Briefcase it leads across analytical quality and presentation sub-scores, and sits just behind Fable 5.1 for rubric-based scoring