Industry developments · Technology and innovation · Global trend ·
Microsoft's research on AI agents at work: workers move faster than their organizations
Microsoft's Work Trend Index surveyed 31,000 knowledge workers in 2025 and 20,000 AI users in 2026. It reports a shift from AI assistants to AI agents, and links the benefits people report more closely to culture, manager support and talent practices than to individual mindset and behaviour.
- Published
- Sources checked
- Sources 3, all primary
- Where Global · Workplace software and knowledge work
A curated item, not a live report: it was checked on 3 November 2026 and does not update by itself. Later events may have changed the picture.
- Knowledge workers surveyed in 31 markets for the 2025 report Source [1]
- 31,000
- Shares of reported AI impact linked to organizational and to individual factors, 2026 Source [2]
- 67% and 32%
- Growth in active agents in Microsoft 365 over a year, as Microsoft reports Source [2]
- 15x
Section 01: What happened?
Sourced facts only, each with the date it refers to and the document that states it.
The development
-
Microsoft's 2025 Work Trend Index drew on a survey of 31,000 workers in 31 countries, labour market data from LinkedIn and Microsoft 365 usage signals. It described a new kind of organization, which it called the Frontier Firm, built around teams of people and AI agents.
Source: [1] Microsoft (WorkLab), company report of 23 April 2025
-
In that survey, 82% of leaders said it was a pivotal year to rethink strategy and operations, and 81% expected AI agents to be moderately or extensively part of their company's AI strategy within 12 to 18 months. 24% said their companies had deployed AI across the whole organization, and 12% were still at the pilot stage.
Source: [1] Microsoft (WorkLab), company report of 23 April 2025
-
45% of leaders named expanding capacity with digital labour as a top priority, second only to upskilling existing staff at 47%, and 33% were considering reducing headcount. The report suggested that the traditional organization chart might give way to teams formed around goals rather than functions.
Source: [1] Microsoft (WorkLab), company report of 23 April 2025
-
The 2025 survey was conducted online by Edelman Data x Intelligence between 6 February and 24 March 2025, with 1,000 workers in each of the 31 markets.
Source: [1] Microsoft (WorkLab), company report of 23 April 2025
-
The 2026 Work Trend Index surveyed 20,000 knowledge workers who use AI, in 10 countries, again through Edelman Data x Intelligence. It reported that organizational factors (culture, manager support and talent practices) accounted for more than twice the reported AI impact of individual factors, 67% against 32%, and stated that the relationship is a statistical association, not a causal effect.
Source: [2] Microsoft (WorkLab), company report of 5 May 2026
-
Only 26% of the AI users surveyed said their leadership was clearly and consistently aligned on AI, and 13% said they were rewarded for redesigning work with AI even when the results fell short. Microsoft reported that the number of active agents in Microsoft 365 had grown fifteenfold in a year.
Source: [2] Microsoft (WorkLab), company report of 5 May 2026
Since then
-
In its annual report, Microsoft said it had introduced new ways of operating designed to follow the work rather than the organization chart, including cross-functional cohorts and squads.
Source: [3] Microsoft Corporation, annual report of 14 October 2026
From here to the discussion question, the page gives the platform's reading of the facts above: the context, the concepts, the choices, the implications and the lessons. It is interpretation, not reporting, and it adds no new facts about the organizations.
Section 02: What is the business context?
The organization, its industry and the situation it was in, read from the facts above.
Microsoft is both the author of this research and a seller of what it describes. It sells AI agents through Microsoft 365, and it reports that the number of active agents there grew fifteenfold in a year. The surveys reach knowledge workers in many countries, 31 markets in 2025 and 10 in 2026, so the context is not one firm or one industry but the way office work is organized almost everywhere.
That work is changing unevenly. In the 2025 survey only 24% of leaders said their company had deployed AI across the whole organization, 12% were still at the pilot stage, and leaders' priorities divided between upskilling existing staff, adding what Microsoft calls digital labour and, for a third of them, reducing headcount. By 2026 the question had moved from whether people use AI to whether their organizations are set up to benefit from it: only 26% of the AI users surveyed saw clear and consistent alignment among their leaders on AI. This is the setting in which the human resource planning choices below are made.
Section 03: Which management concepts are relevant?
Where this development meets the IB course, and where it reaches beyond it, with links to the subtopics, terms, frameworks and tools that explain it.
In the IB course
-
2.1 Introduction to human resource management
Human resource planning, reasons for resistance to change, and strategies to reduce its impact.
-
2.2 Organizational structure HL content
Project-based organization and teams formed around goals rather than functions.
-
2.5 Organizational (corporate) culture HL
Organizational culture as the factor most closely linked to the benefits people report.
-
Evaluating secondary research; who ran the survey, who was asked, and who paid for it.
-
5.9 Management information systems HL
Artificial intelligence and big data, and the ethics of using data about how employees work.
Key terms
Toolkit and frameworks
-
Force field analysis IB toolkit HL
Weigh the forces driving and restraining a move to AI agents in one organization.
-
STEEPLE analysis IB toolkit
Technological and social change in the external environment, and the ethical questions it brings.
Beyond the course
Source: EnrichmentFor undergraduate, MBA and professional readers. These ideas are not part of the IB course and are not assessed by it.
-
Leading a move to AI agents as a change programme: urgency, a guiding coalition, short-term wins, and new ways of working anchored in the culture.
-
Strategy, structure and systems change with the technology; shared values, style, staff and skills decide whether the change holds.
Terms beyond the course
Work it through on the platform
-
Plan the introduction of AI agents in one team: the forces for and against, the people affected and the steps.
-
Work out the staff a team will need when some tasks move to agents, and the gap to fill by training or hiring.
Section 04: What choices did the organization face?
The realistic options, with the case for and against each. Where the organization's choice is known from the facts above, it is marked.
-
Wait until the technology matures
Avoids early mistakes and costs, but the 2026 report argues that firms which learn early pull ahead, and staff may adopt tools anyway without rules to guide them.
-
Run small pilots in a few teams
Low risk and good evidence, but pilots alone can leave culture, management and talent practices unchanged, and those are what the 2026 report links most closely to results.
-
Redesign work across the whole organization
Captures the largest gains if it works, but is the hardest change to manage and the most exposed to resistance and to fears about jobs.
-
Invest first in managers, training and incentives
Follows the 2026 finding that organizational factors matter most, but takes time and money before results show.
Section 05: What were the implications?
What followed, or may follow, for the organization, its stakeholders and its industry.
For organizations, the 2026 finding moves the problem. If the benefits people report are linked more to culture, manager support and talent practices than to individual mindset and behaviour, then buying AI tools is the easy part, and the harder part is change management: aligning leaders, preparing managers and changing how people are developed and rewarded. Only 13% of the AI users surveyed said they were rewarded for redesigning work with AI even when the results fell short, which suggests that most reward systems still discourage the experiments that adoption needs.
For employees the implications are mixed. A third of leaders in 2025 were considering reducing headcount, and teams formed around goals rather than functions would change reporting lines and roles; Microsoft says it has begun to organize some of its own work in cross-functional cohorts and squads. For anyone using the research, the implication is to weigh it as evidence: a supplier's survey, self-reported, of people who already use AI, reporting an association rather than a cause. Microsoft says the last of these itself.
Section 06: What can a student or manager learn?
- Technology changes outcomes only when the organization changes with it, in its culture, its managers and the way it develops and rewards people.
- Read research for who produced it and who was asked. A supplier's survey of its users can be useful evidence without being neutral evidence.
- An association is not a cause. The report says so itself, and so should anyone who uses its figures.
- For managers, decide what people will stop doing, not only what AI will start doing.
Section 07: Discussion question
With reference to the Work Trend Index findings, examine the human resource challenges a business faces when it introduces AI agents into its workforce.
Command term: Examine (AO3). The platform's own question for discussion; it is not an IB examination question and has no official markscheme.
Section 08: The limits of the evidence
What the sources do not show, so that no one takes the analysis for more than it is.
- Both reports are published by Microsoft, which sells the AI products they discuss.
- The 2026 survey included only workers who already use AI, so it cannot show the views of those who do not.
- The answers are self-reported, and the 2026 report gives two different end dates for its fieldwork, 7 April 2026 in its methodology and 20 April 2026 beside its charts.
Section 09: Sources
Every source was opened and read on the date shown. Each entry says which facts on this page it supports.
-
[1] Company report · Primary source
2025: The year the Frontier Firm is born (external site)
https://www.microsoft.com/en-us/worklab/work-trend-index/2025-the-year-the-frontier-firm-is-born
Supports
- The data behind the 2025 report and its description of the Frontier Firm.
- The shares of leaders on strategy, agents, organization-wide deployment and pilots.
- The priorities for digital labour, upskilling and headcount, and the idea of teams formed around goals.
- Who ran the 2025 survey, how, when and with how many workers in each market.
-
[2] Company report · Primary source
Agents, human agency, and the opportunity for every organization (external site)
https://www.microsoft.com/en-us/worklab/work-trend-index/agents-human-agency-and-the-opportunity-for-every-organization
Supports
- The 2026 survey of 20,000 AI users in 10 countries and who ran it.
- The 67% and 32% shares for organizational and individual factors, and the statement that this is an association, not a causal effect.
- The shares on leadership alignment and rewards for redesigning work, and the fifteenfold growth in active agents.
- The two different fieldwork end dates given in the report.
-
[3] Annual report · Primary source
Microsoft 2026 Annual Report (external site)
https://www.microsoft.com/investor/reports/ar26/index.html
Supports
- Microsoft's statement, in the shareholder letter, that it introduced ways of operating that follow the work rather than the organization chart.
The shareholder letter that opens the report is dated 14 October 2026.
Arjun Agrawal | The Business Brief, https://the-business-brief-dun.vercel.app/intelligence/ai-agents-at-work/. Original teaching material, not an IB publication. The organizations in the teaching cases are fictional unless a case says otherwise.