Companies that use AI openly and let people change how they work seem to get more value from it than companies that just install the tool.
What distinguishes AI high performers from other organizations using AI?
This explores what separates organizations that get real value from AI from those that adopt it without much to show. The retrieved notes don't include a direct study of 'high performers', so this answer pieces together the factors the corpus suggests make the difference.
This explores what separates organizations that get real value from AI from those that adopt it without much to show. One caveat first: none of these retrievals profiles 'AI high performers' directly. Instead, the collection keeps finding gaps between having AI and benefiting from it. Read together, those gaps suggest what the organizations that close them are doing differently.
The first gap is visibility. Most people using AI at work are hiding it. In Anthropic's interviews, most workers reported saving time, yet about 70% mentioned the stigma of using AI at work Why do workers hide productivity gains from AI use?. Experiments show why: people expect colleagues to see them as less competent and less diligent if they admit to using AI Do people fear judgment when they use AI at work?. An organization can't redesign work around gains it can't see, so one plausible difference is simply making AI use safe to talk about. That turns private shortcuts into shared practice. Microsoft's push toward 'collective productivity' points the same way: the next gains come from AI built around teams, not individual tools Can AI boost how teams work together?.
The second gap is between confidence and results. In WalkMe's survey, 90% of workers felt confident with AI, but only 25% said it worked on the first try. Half had spent more time using AI on a task than doing it by hand Why do workers feel confident with AI but get poor results?. Even when the AI improves, people working with it capture only about half of the improvement Why does assisted accuracy capture only half the LLM gain?. Executives also tend to report bigger productivity gains than can be measured, partly because revenue lags behind operational changes Do AI productivity gains feel larger than they actually measure?. Organizations that measure real outcomes rather than enthusiasm are more likely to notice where the value leaks out.
The third gap is about where in the work AI lands. Narayanan and Kapoor argue that AI mostly shrinks the middle 'execute' layer of knowledge work. Deciding what to do and delivering it to people stay the same size or grow Does AI really compress all layers of knowledge work equally?. Benchmarks hide this: agents win contests but struggle with long, messy professional workflows Why do agent benchmarks not predict real economic value?. They also outscore experts on 2-hour tasks but fall behind over 32 hours When do AI agents outperform human research experts?. The organizations that pull ahead are probably the ones that test AI on their own long, real tasks instead of trusting leaderboards Do automated benchmarks hide what frontier AI systems can really do?. They also put people where judgment matters, since choosing which details matter is still a human skill Can AI distinguish which differences actually matter?.
The surprising takeaway: the edge seems to come less from better models than from better organizational habits. That means making AI use visible, measuring real outcomes, and moving people toward deciding and delivering, where AI doesn't compress the work. If you want hard numbers on who the high performers are, this part of the collection doesn't have them yet.
Sources 11 notes
In a 1,250-person interview study, 86% of general workers and 97% of creatives said AI saved them time, yet 69% of workers and 70% of creatives mentioned workplace stigma or managing peer judgment around their AI use.
Across four experiments with 4,439 participants, people using AI expected others to judge them as less competent and diligent, and reported lower willingness to disclose AI use to managers and colleagues. The gap suggests a social cost that users foresee and act on.
Microsoft's 2025 report argues the next AI frontier is collective productivity, requiring systems built around shared goals and collaboration norms rather than individual tools. The claim frames this as a deliberate design mandate, though the excerpt provides no empirical evidence of collective-productivity gains.
WalkMe's survey of 2,037 US workers found 90% feel confident using AI, but only 25% report it works on first try and 50% spent more time using AI than doing tasks manually. The gap widened most among younger workers, suggesting overestimation of skill.
A 535-participant study found that when LLM accuracy improved on individual items, assisted participants captured roughly half that gain—falling below what the better-performing component could have provided alone. This shows complementarity creates potential but does not guarantee synergy.
Show all 11 sources
A survey of 750 executives found that perceived AI productivity gains exceed measured ones, likely because revenue lags operational improvements. Effects concentrate in high-skill services and finance, with labor reallocating rather than shrinking overall.
Narayanan and Kapoor argue AI narrows only the middle execution layer of knowledge work while decide and deliver layers persist or grow. Translation and legal work show stable or expanding employment despite AI gains, suggesting task-level compression doesn't shrink occupational demand.
ALE's analysis of 960 real occupational workflows shows agents excel at abstract contests but fail long-horizon professional tasks. The gap is not model capability but benchmark design—the field optimizes what it measures, and it has measured contests rather than work.
METR's RE-Bench found AI agents score 4× higher than expert humans at 2-hour budgets but humans narrowly exceed agents at 8 hours and lead 2× at 32 hours, suggesting agents hit scaling plateaus while humans improve with extended effort.
Automated benchmarks both overstate and understate capability by privileging precisely-specified, auto-gradable tasks. Open-world evaluations of long-horizon messy tasks through qualitative log analysis—with cost explicitly reported—correct these distortions and catch emerging capabilities earlier.
Experts observe by choosing which differences matter (qualitative judgment); AI finds patterns and probabilities (quantitative). AI generates text from prompts without observing context, audience needs, or knowledge states—producing fabrication that mimics observation's form without its epistemic process.
Papers this line draws on 8
The research behind the notes this line reads — ranked by how closely each paper relates.
- The Rise of Industrial AI in America: Microfoundations of the Productivity J-curve(s)
- What 81,000 people told us about the economics of AI
- Measuring AI Ability to Complete Long Software Tasks
- PostTrainBench: Can LLM Agents Automate LLM Post-Training?
- Beyond Productivity: Measuring the Real Value of AI
- Agents' Last Exam
- RE-Bench: Evaluating frontier AI R&D capabilities of language model agents against human experts
- Evidence of a social evaluation penalty for using AI