Build
Connect & operate
Design & teams
Start hereScope a build in one callBring a spec, a wireframe, or a paragraph. You leave with an architecture, a timeline, and a number.Request a scoping call
AI software
LLM & data systems
Vibe coding
Ready to ship?Put AI where the work isAgents, RAG, and private LLMs wired into the systems your team already uses — not a chatbot bolted to a homepage.Discuss an AI project
Domain firstWe learn your workflow before we model itRegulated, operational, or high-volume — the constraints belong in the schema, not in a training doc.Talk about your domain
Graphic · AI at Work · 2026

How People Use AI at Work (2026)

The Thinking Share: nearly half of what workers ask Microsoft 365 Copilot to help with is analyzing, reasoning and deciding, not writing. The single biggest use is making decisions and solving problems, at 27.5%. Meanwhile, 32% of U.S. employees now use AI at work a few times a week or more.

Updated Nov 20, 20266 charts · 4 key stats7 primary sources
Fig. 1 · The Thinking Share

Nearly half of Copilot requests at work are for thinking, not producing

How people use AI at work: share of Microsoft 365 Copilot user goals by work-activity category (left) and by activity (right), one week of February 2026. Blue: analyzing, reasoning and deciding.

Flows from Analyzing, reasoning & decidingHover or tap a bar to trace its flows
Activity categoryWork activityAnalyzing, reasoning & deciding49%Interacting with others19.1%Producing work16.7%Information gathering15.2%Making decisions & solving problems27.5%Analyzing data or information5.5%Thinking creatively4.9%Processing information3.1%Judging qualities of things or people2.8%Other reasoning activities5.2%Communicating with supervisors & peers8.4%Interpreting information for others4.5%Other people activities6.3%Documenting & recording information11.7%Working with computers4.7%Handling & moving objects0.3%Getting information13%Other information gathering2.2%

Values are Microsoft's published shares (×100) of classified Copilot user goals, not time spent or session counts. Each chat is classified by user intent against the O*NET work-activity taxonomy; a chat with several goals is counted fractionally. "Other" rows are dev.co sums of the smaller activities in each category. Both columns total 100%.

Copilot user goals that are analyzing, reasoning or deciding
49%
vs. 17% producing work (Feb 2026)
Most common use of AI at work: making decisions & solving problems
27.5%
of Copilot user goals; next is getting information, 13.0%
U.S. employees who use AI at work a few times a week or more, Q3 2026
32%
up from 11% in Q2 2023; 16% use it daily
Fig. 2

Frequent AI use at work has nearly tripled since 2023

AI use at work statistics: share of U.S. employees who use AI in their role, by frequency, Q2 2023 to Q3 2026, %

020%40%60%Q2 2023Q2 2024Q2 2025Q2 2026Total53% in Q3 2026Frequent32% in Q3 2026Daily16% in Q3 2026

Gallup labels each wave by its survey month (May 2023 through August 2026); quarters here follow Gallup's industry file. Gallup's published series has one point a year through Q2 2025, then one per quarter. The 2.9× rise (11% to 32%) is dev.co's calculation.

Fig. 3

Tech workers lead daily AI use at 44%

AI use at work by industry: share of U.S. employees who use AI daily, Q3 2026, %

Technology44%Finance29%Professional services24%Healthcare17%College or university16%Retail13%Manufacturing12%Social services11%Government11%K-12 education8%

Gallup's 10 reported industries; social services is Gallup's community/social services, government is government or public policy. In technology, 67% use AI frequently and 84% use it at least a few times a year.

Fig. 4

AI frees up time for searching and writing, not for meetings

Does AI save time at work? Share of frequent U.S. AI users who spend less vs. more time on each activity since they began using AI, Aug 2026, %

More time Less time Gap (Less time − More time)
020%40%60%Gap↑ More say less time↓ More say more timeSearching for or gathering information8%52%+44 ptsWriting and drafting7%47%+40 ptsAnalyzing data11%38%+27 ptsManual or administrative tasks7%29%+22 ptsCreative or idea-generation work13%29%+16 ptsDeep thinking or problem solving13%22%+9 ptsPlanning and prioritizing11%16%+5 ptsCollaborating with coworkers9%13%+4 ptsReviewing and correcting work21%23%+2 ptsWorking with customers or clients8%10%+2 ptsAttending meetings8%10%−2 ptsLearning new skills or tools18%21%−3 pts

The rest of each row said "about the same" (from 40% for searching to 83% for meetings). Frequent users use AI a few times a week or more. Rows may not sum to 100% because of Gallup's rounding.

SourceGallup4
Fig. 5

Leaders in desk jobs are the heaviest AI users

Share of U.S. employees who use AI at work a few times a week or more, by job type, Aug 2026, %

Desk jobs: leaders57%Desk jobs: all49%Desk jobs: staff46%All U.S. employees32%On-site jobs22%

Desk jobs are Gallup's remote-capable jobs: desk or computer jobs that could be done remotely, wherever people actually work. Staff are individual contributors. Daily use: 32% of remote-capable leaders, 25% of remote-capable individual contributors, 26% of all remote-capable workers, 10% of on-site workers.

SourceGallup4
Fig. 6

Only 19% of AI users have both the skills and the support

AI users by Microsoft's readiness zone (personal AI skill vs. employer readiness), 10 countries, Feb–Apr 2026, respondents

19.0%50.0%15.8%10.2%
  • Frontier3,22619.0%
  • Emergent8,48550.0%
  • Stalled2,68515.8%
  • Blocked agency1,72610.2%
  • Unclaimed capacity8495.0%
  • AI users scored16,971100%

Frontier: above the median on both personal AI capability and employer readiness. Blocked agency: high personal, low employer. Unclaimed capacity: the reverse. Stalled: below the median on both. Emergent: everyone near the middle. Scored on 16,971 of 20,000 knowledge workers who use AI.

Free to republish

Add This Graphic to Your Website

Paste one snippet and every chart, stat and citation above appears on your page. It stays interactive and responsive, and it updates automatically when we revise the data.

  • All sources travel with every card
  • Resizes to fit any column width
  • Light or dark theme, full set or a single card
  • Free with attribution, which is built in
Share this graphic
<div class="graphic-embed" data-graphic="how-people-use-ai-at-work">
  <a href="https://dev.co/graphics/how-people-use-ai-at-work">How People Use AI at Work (2026)</a> by DEV.co
</div>
<script async src="https://dev.co/embed.js"></script>
Auto-resizes to fit its content. Works on WordPress, Webflow, Squarespace, Ghost and plain HTML.
Preview

How are people using AI at work?

Mostly to think. In one week of February 2026, 49% of the goals people brought to Microsoft 365 Copilot were analyzing, reasoning and deciding, according to Microsoft's 2026 Work Trend Index. Working with other people made up 19%, producing work 17% and gathering information 15%. Microsoft based the split on more than 100,000 Copilot chats, each classified against the U.S. government's O*NET list of work activities.

The Thinking Share, Figure 1, follows each of those four categories to the activities inside it. The analyzing, reasoning and deciding band is drawn in blue. Writing still matters, since documenting and recording information is 11.7% of all goals, but it is not the main job people give the assistant.

What is the most common use of AI at work?

Making decisions and solving problems is the most common use, at 27.5% of Copilot user goals. That is more than twice the next activity, getting information (13.0%), and more than documenting and recording information (11.7%) and communicating with supervisors and peers (8.4%) combined. Analyzing data (5.5%) and thinking creatively (4.9%) follow.

Workers in Microsoft's survey of 20,000 AI users see the same shift. Asked which human skills matter more as AI takes on more work, 50% named quality control of AI output and 46% named critical thinking. And 86% said they treat AI output as a starting point, not a final answer.

How many people use AI at work?

About half of U.S. employees (53%) used AI in their role at least a few times a year in the third quarter of 2026, according to Gallup's survey of 21,648 workers. Frequent use, a few times a week or more, reached 32%, up from 11% in the second quarter of 2023. Daily use reached 16%, up from 4%.

Use depends heavily on the job. Nearly half (49%) of workers in remote-capable desk jobs use AI frequently, compared with 22% of workers whose jobs must be done on site. Among remote-capable workers, 57% of leaders use AI frequently, compared with 46% of individual contributors.

Does AI save time at work?

For some tasks, yes. Among frequent users, 52% spend less time searching for or gathering information since they began using AI, and 47% spend less time writing and drafting, Gallup found. On the other hand, 83% spend about the same time in meetings, 82% the same time with customers and 78% the same time collaborating with coworkers.

Two activities moved both ways. About one in five frequent users now spend more time reviewing and correcting work (21%) and learning new skills and tools (21%), while 23% and 18% spend less. Most frequent users (65%) say their work takes about the same mental effort as before.

Which industries use AI at work the most?

Technology leads by a wide margin: 44% of U.S. tech workers used AI daily in the third quarter of 2026, compared with 29% in finance and 24% in professional services. In Gallup's seven other industries, daily use ranges from 8% (K-12 education) to 17% (healthcare).

Readiness, not access, is the bigger gap inside companies. Microsoft scored AI users on their own AI capability and on their employer's readiness to support it. Only 19% scored above the median on both, the group Microsoft calls the Frontier, while 10% have the skills but not the support.

Methodology & caveats

Figure 1 and the first two stat tiles use Microsoft 365 Copilot telemetry from one week in February 2026, published as chart data with Microsoft's 2026 Work Trend Index. Percentages are each activity's share of classified user goals, not time spent or number of sessions, and a chat with several goals is counted fractionally. The report's methodology describes this sample as 105,000 commercial chats from North America (excluding education), while the chart data file describes its scope as all segments worldwide; we follow the report text. The data covers one product's users, so it describes how people use Copilot, not all AI tools. Values shown are Microsoft's published shares multiplied by 100. The "other" rows in Figure 1 are dev.co sums of smaller activities.

Microsoft's survey figures (Figure 6 and the skill and starting-point shares) come from an online survey by Edelman Data x Intelligence of 20,000 knowledge workers who use AI at work, in 10 countries, from February 18 to April 7, 2026. People who never use AI were screened out, so these are shares of AI users, not of all workers. The readiness zones were scored on 16,971 respondents.

Gallup's figures come from its quarterly web surveys of U.S. employees aged 18 and older drawn from the probability-based Gallup Panel. The latest wave ran August 5–19, 2026, with 21,648 respondents and a margin of error of ±0.9 percentage points. Gallup's published series has one point a year from 2023 to 2025 and one per quarter after that. Industry values are Gallup's Q3 2026 figures, which carry larger margins of error than the national total. The 2.9× rise in frequent use and the net figures in Figure 4 are our calculations. Microsoft and Gallup measure different populations, so their numbers should not be compared directly.

Sources

  1. [1]
    Microsoft. 2026 Work Trend Index chart data: share of Copilot user goals by O*NET work activity (treemap, February 2026)https://assets-c4akfrf5b4d3f4b7.z01.azurefd.net/assets/2026/04/DV1_updated050126_V3_69f4c13474f74.json
  2. [2]
    Microsoft. 2026 Work Trend Index Annual Report (PDF): p. 8 chart and methodology, p. 22–23https://assets-c4akfrf5b4d3f4b7.z01.azurefd.net/assets/2026/09/2026_Work_Trend_Index_Annual_Report_090326_6a99e3f096b6e.pdf
  3. [3]
    Microsoft. 2026 Work Trend Index Annual Report: Agents, human agency, and the opportunity for every organizationhttps://www.microsoft.com/en-us/worklab/work-trend-index/agents-human-agency-and-the-opportunity-for-every-organization
  4. [4]
    Gallup. How AI Is Changing the Way Employees Spend Time at Work (16 Nov 2026)https://www.gallup.com/workplace/716585/ai-changing-employees-spend-time-work.aspx
  5. [5]
    Gallup. Indicator: Artificial Intelligence (AI use at work, data as of August 2026)https://www.gallup.com/699797/indicator-artificial-intelligence.aspx
  6. [6]
    Gallup. AI Use Trends by Industry (spreadsheet, Q2 2023 to Q3 2026)https://www.gallup.com/file/education/715805/AI-Use-Trends-by-Industry.xlsx
  7. [7]
    Microsoft. 2026 Work Trend Index chart data: organization vs. employee AI readiness zoneshttps://assets-c4akfrf5b4d3f4b7.z01.azurefd.net/assets/2026/04/dv3_dynamicContent_69f65be171891.json
Cite this graphicDEV.co. “How People Use AI at Work (2026).” DEV.co Graphics, updated 2026-11-20. https://dev.co/graphics/how-people-use-ai-at-work