The Future of Work Report 2026: AI Agents Will Augment 80% of Knowledge Worker Tasks by 2028

The Definitive Forecast for AI's Impact on Work

The World Economic Forum (WEF) and Gartner have jointly published "The Future of Work 2026: The Agentic AI Transformation," a comprehensive 247-page report that represents the most rigorous analysis to date of how AI agents will reshape knowledge work over the next two years. Released on March 13, 2026, the report draws on surveys of 12,000 enterprises across 45 countries, interviews with 800 C-suite executives, analysis of 1.2 billion job postings, and input from 150 academic researchers.

The central finding is stark: by 2028, AI agents will augment approximately 80% of tasks currently performed by knowledge workers. This does not mean 80% of knowledge workers will lose their jobs — the report is careful to distinguish between task-level augmentation and job-level displacement. But it does mean that virtually every knowledge worker's daily workflow will be fundamentally reshaped by AI agents within 24 months.

"We have not seen a workforce transformation of this magnitude since the introduction of the personal computer," said Saadia Zahidi, Managing Director of the WEF and head of the Centre for the New Economy and Society. "The PC changed what knowledge workers could do. AI agents will change what knowledge workers need to do."

Key Predictions

Prediction 1: 80% Task Augmentation by 2028

The 80% figure refers to the percentage of discrete tasks within knowledge worker roles that will be either automated (performed entirely by AI agents) or augmented (performed by humans with AI agent assistance) by 2028. The report categorizes knowledge worker tasks into four quadrants.

flowchart LR
    REL(["Release of<br/>The Future of Work<br/>Report 2026"])
    NEW1["What's new<br/>flagship feature 1"]
    NEW2["What's new<br/>flagship feature 2"]
    NEW3["What's new<br/>flagship feature 3"]
    BREAK{"Breaking<br/>changes?"}
    MIG["Migration steps"]
    UPG(["Upgrade now"])
    WAIT(["Pin current,<br/>upgrade later"])
    REL --> NEW1
    REL --> NEW2
    REL --> NEW3
    NEW1 --> BREAK
    NEW2 --> BREAK
    NEW3 --> BREAK
    BREAK -->|Yes| MIG --> UPG
    BREAK -->|No| UPG
    BREAK -->|Risk averse| WAIT
    style REL fill:#4f46e5,stroke:#4338ca,color:#fff
    style BREAK fill:#f59e0b,stroke:#d97706,color:#1f2937
    style UPG fill:#059669,stroke:#047857,color:#fff
    style WAIT fill:#0ea5e9,stroke:#0369a1,color:#fff

Fully automatable tasks (estimated 25% of current knowledge worker tasks) include data entry and transfer between systems, routine email responses and scheduling, standard report generation, basic data analysis and summarization, and first-tier customer and employee support inquiries. These tasks will be handled entirely by AI agents, requiring human oversight only for exception handling.

Agent-augmented tasks (estimated 55% of current tasks) involve research and information synthesis, document drafting and editing, complex analysis and recommendation development, project coordination and status tracking, and decision preparation where the agent does the analysis and the human makes the call. For these tasks, AI agents will handle the mechanical and analytical components while humans provide judgment, creativity, and stakeholder management.

Human-essential tasks (estimated 20% of current tasks) remain squarely in human territory. These include relationship building and trust establishment, ethical judgment in ambiguous situations, creative ideation and strategic vision, cross-cultural negotiation and diplomacy, and crisis leadership and stakeholder communication under pressure.

Prediction 2: The "Agent Dividend" Will Add $4.3 Trillion to Global GDP by 2028

The report estimates that AI agent-driven productivity gains will add $4.3 trillion to global GDP by 2028, distributed unevenly across regions and industries. The United States is projected to capture $1.4 trillion of this value, followed by China ($890 billion), the European Union ($720 billion), and Japan ($340 billion).

"The agent dividend is not automatic," cautioned Daryl Plummer, Distinguished VP Analyst at Gartner and co-author of the report. "It accrues to organizations and economies that invest in AI agent infrastructure, workforce reskilling, and organizational redesign. Countries and companies that delay will not simply miss the dividend — they will lose competitive ground to those that capture it."

The GDP impact is driven primarily by productivity gains (estimated to account for 60% of the value), new product and service creation enabled by AI agents (25%), and reduced operational waste and error costs (15%).

Prediction 3: 40% of Current Job Roles Will Be "Redesigned" by 2028

The report projects that 40% of existing knowledge worker job descriptions will be substantially redesigned to account for AI agent capabilities. This is distinct from job elimination — the report estimates net job losses of 5-8% in knowledge work, far less than the "50% of jobs will disappear" predictions that dominated headlines in 2023-2024.

The redesign will follow a consistent pattern. Routine and analytical components of roles will be delegated to AI agents. Human workers will be "elevated" to focus on judgment, creativity, relationship, and leadership components. New responsibilities will emerge around agent management, output quality assurance, and human-AI workflow design.

The report provides detailed role transformation projections for 50 common knowledge worker roles. Some examples from the report illustrate the pattern.

Financial analysts currently spend 60% of their time gathering data, building models, and preparing reports, with 40% on interpretation, client communication, and strategic recommendation. By 2028, the report projects this will shift to 15% on overseeing AI agent data gathering and model building, with 85% on interpretation, client relationships, and strategic advice.

Marketing managers currently allocate roughly 50% of their time to campaign execution, content production, and performance reporting. By 2028, AI agents will handle 80% of these execution tasks, freeing marketing managers to focus on brand strategy, creative direction, and customer insight — areas where human creativity and judgment remain essential.

Software engineers, already among the most AI-augmented knowledge workers, are projected to see their coding time reduced by 70% through AI pair programming agents. The role will shift toward architecture design, code review, system reliability, and translating business requirements into technical specifications.

Prediction 4: "Agent Management" Becomes a Core Competency

The report introduces the concept of "agent management" as a new core competency that will be as essential to knowledge workers as digital literacy became in the 2000s. Agent management encompasses the ability to effectively instruct and configure AI agents, evaluate and verify agent outputs, design workflows that optimally combine human and agent capabilities, troubleshoot agent failures and escalation scenarios, and maintain security and compliance standards in agent-assisted processes.

Gartner projects that by 2028, 65% of enterprise job postings will list AI agent management skills as a requirement, up from 4% in early 2026.

"The workers who thrive in the agentic era will not be those who resist AI agents or those who defer to them unconditionally," said Zahidi. "They will be those who develop the judgment to know when to trust the agent, when to override it, and how to combine human insight with AI capability for results neither could achieve alone."

Prediction 5: The "Hollow Middle" Risk

The report identifies a significant structural risk it calls the "hollow middle" — the potential elimination of mid-level knowledge worker roles that serve primarily as information conduits between senior leadership and front-line workers. These roles, which include many middle management positions, analyst roles, and coordination functions, are disproportionately vulnerable to AI agent augmentation because their primary value is information processing and communication rather than judgment or creativity.

"The hollow middle is not inevitable," the report emphasizes. "But it requires deliberate organizational redesign to prevent. Companies that simply layer AI agents on top of existing organizational structures will find that middle layers become redundant. Companies that proactively redesign their organizations around human-AI collaboration will create new middle-tier roles focused on agent orchestration, quality assurance, and cross-functional coordination."

Industry-Specific Analysis

The report provides detailed transformation timelines for 14 industry sectors. Key highlights include the following projections.

Financial services is the sector most advanced in AI agent adoption, with 67% of major financial institutions already deploying AI agents in at least one business function. The report projects that by 2028, AI agents will handle 90% of routine banking interactions, 75% of insurance claims processing, and 60% of financial advisory preparation work.

Healthcare is projected to see the most transformative impact, though adoption timelines are longer due to regulatory requirements. AI agents are expected to handle 70% of administrative tasks (scheduling, billing, insurance verification), 50% of diagnostic triage, and 30% of patient communication by 2028.

Legal services face significant disruption, with AI agents projected to automate 65% of document review, 55% of legal research, and 40% of contract drafting. The report notes that this will not reduce the number of lawyers but will fundamentally change what lawyers spend their time doing.

Education will see AI agents serving as personalized tutors for 60% of K-12 students in developed countries by 2028, with the teacher's role shifting from content delivery to mentorship, emotional support, and creative facilitation.

Workforce Transition Recommendations

The report's policy recommendations section calls for unprecedented coordination between governments, educational institutions, and the private sector.

For governments, the report recommends establishing national AI agent literacy programs, updating labor regulations to address AI agent-augmented work, creating portable benefits systems for workers transitioning between roles, and investing in research on long-term social impacts of agent augmentation.

For enterprises, the recommendations include beginning workforce redesign now rather than waiting for full AI agent maturity, investing in reskilling programs focused on agent management competencies, adopting graduated autonomy models where AI agents start with limited authority and earn expanded capabilities as trust is established, and creating dedicated human-AI collaboration design teams.

For educational institutions, the report advocates integrating AI agent literacy into core curricula at all levels, redesigning professional education to emphasize judgment, creativity, and human skills that complement AI capabilities, and establishing research programs focused on human-AI collaboration effectiveness.

The Bigger Picture

The WEF-Gartner report represents a maturation in the discourse around AI and the future of work. Gone are the binary predictions of mass unemployment or AI-as-panacea. In their place is a nuanced analysis that acknowledges both the transformative potential and the significant risks of agentic AI in the workplace.

The 80% task augmentation projection is both reassuring and sobering. Reassuring because it frames AI agents as tools that enhance human capability rather than replace human workers wholesale. Sobering because the speed and scale of the transformation — fundamentally reshaping 80% of knowledge worker tasks within 24 months — requires organizational and societal adaptation at a pace that history suggests is achievable but not guaranteed.

"The agentic AI transformation is happening whether organizations prepare for it or not," concluded Plummer. "The question is not whether to adapt, but whether to adapt proactively and capture the value, or reactively and absorb the disruption. There is no third option of standing still."

For the estimated 1.1 billion knowledge workers worldwide, the message is clear: the daily experience of work is about to change more profoundly and more rapidly than at any point in the modern era. How that change unfolds — whether it creates broadly shared prosperity or concentrated disruption — depends on decisions being made right now by leaders in government, business, and education.

Sources

• World Economic Forum and Gartner, "The Future of Work 2026: The Agentic AI Transformation," March 2026
• Financial Times, "WEF-Gartner Report: AI Agents to Reshape 80% of Knowledge Work by 2028," March 2026
• The Economist, "The Agentic AI Revolution in the Workplace: Hype vs. Reality," March 2026
• Harvard Business Review, "Preparing Your Organization for the Agent-Augmented Workforce," March 2026
• Brookings Institution, "Policy Responses to AI Agent Workforce Disruption," March 2026