Labor Intelligence · AI Displacement · 2026 Data
The Complete Map of
Jobs at Risk
Across Every Sector
An evidence-based, sector-by-sector breakdown of which specific jobs AI and automation will displace — and which ones will survive. As granular as the data allows. Updated daily.
Global AI Job Displacement — Live Estimates
Global jobs at significant risk
300M
Over next 5–10 years
US jobs at risk
47M
High automation susceptibility
New jobs to be created
97M
Emerging AI-era roles
Jobs augmented vs replaced
65%
Will be transformed, not eliminated
Net displacement: −83M net displacement · Sources: BLS 2024, Oxford Frey & Osborne, Goldman Sachs AI Research, McKinsey Global Institute, WEF Future of Jobs 2025, OECD Employment Outlook, PwC AI Jobs Barometer
First principles — what does automation actually replace?
Start from the bottom. A job is a bundle of tasks. Some tasks are routine — they follow predictable rules, operate on structured inputs, and produce measurable outputs. Other tasks are novel — they require judgment, physical presence, emotional attunement, or accountability that cannot be pre-programmed. AI is exceptionally good at the first category and structurally incapable of the second. The risk to any job is therefore a function of what proportion of its tasks are routine vs. novel. A radiologist reading a standard chest X-ray is doing a routine task. A radiologist deciding how to communicate a terminal diagnosis to a patient is doing a novel one. AI will compress the first. It cannot touch the second.
The strategic question is not "will my job be replaced?" It is "what fraction of my job is routine tasks — and what am I doing to shift that fraction?"
How to Read This Analysis
HIGH RISK
Strong evidence of automation in the next 3–7 years. Multiple production deployments already underway.
MEDIUM RISK
Significant automation pressure over 5–10 years. AI augments today; replacement likely in next cycle.
PROTECTED
Physical presence, human judgment, creative authorship, or legal requirement protects these roles.
Risk percentages represent the share of current role holders facing significant disruption based on task automation susceptibility. They do not mean immediate elimination — they represent the probability of substantial role transformation or headcount reduction within the stated timeline. Sources: McKinsey [1], Oxford [2], WEF Future of Jobs 2025 [4].
Sector-by-Sector Breakdown — Click to Expand
Jobs That Will Not Be Impacted
Across all sectors, certain job categories have structural protection that AI cannot currently overcome — physical dexterity in unstructured environments, legally mandated human oversight, human therapeutic relationships, creative originality, and sovereign decision-making.
Mental Health & Therapy
Human therapeutic alliance is the treatment mechanism — cannot be replicated by AI.
Trades & Physical Infrastructure
Unstructured physical environments, dexterous problem-solving, and code-governed work AI cannot yet perform.
Education (High-Touch)
Adaptive human relationship with learners — especially those with special needs — is irreplaceable.
Creative & Cultural
Human origin and authentic cultural production retain economic and social premium that AI-generated content cannot capture.
Government & Policy
Democratic accountability, security clearance requirements, and sovereign decision-making require humans by law or political necessity.
Social Work & Community
Trust-based human relationships in vulnerable contexts — legal and ethical barriers to AI delegation.
Physical Sciences Research
Physical world data collection and hypothesis-driven experimentation in unstructured environments.
Spiritual & Religious
Human meaning-making and spiritual care — no substitute exists or is culturally acceptable.
The Big Picture — What the Research Actually Says
McKinsey Global Institute [1]
Up to 300M jobs globally could be displaced by 2030–2035. But 60% of those displaced will transition to new roles — not permanently exit the workforce. The $13T productivity gain from AI is projected to create 97M+ new roles.
Oxford University — Frey & Osborne [2]
47% of US jobs are at high risk of automation. But crucially: jobs within the same occupation vary dramatically. A radiologist reading routine chest X-rays is high-risk. A radiologist managing complex multi-modality cancer staging is not.
Goldman Sachs [3]
Generative AI could automate 25–50% of current work tasks. The report explicitly notes: most jobs will be transformed, not eliminated. 65% of occupations will see AI augmentation before replacement.
WEF Future of Jobs 2025 [4]
85M jobs displaced by 2025 (actual), 97M new roles created. Net positive — but the mismatch between where jobs are lost (clerical, routine) and created (AI, green tech, care) is the central workforce challenge of the decade.
The Net Honest Verdict
The labor market disruption from AI is real, large, and uneven. It is not a uniform wave — it is a targeted elimination of tasks, not jobs. Roles that survive are those where humans provide something AI structurally cannot: physical presence in unstructured environments, genuine emotional relationship, legally mandated judgment, creative cultural authenticity, or sovereign accountability.
The workers most at risk are not the ones doing the most complex jobs — they are the ones doing the most routine jobs: routine document review, routine data entry, routine visual inspection, routine dispensing. AI handles routine far better than it handles novel.
The strategic career question of the next decade: are you paid primarily for what you know, or what you do with that knowledge in unpredictable, human, physical contexts? The former is at risk. The latter is not.
Start a Critical Discussion
These questions don't have consensus answers. Share one to LinkedIn or X and see what your network actually thinks.
"47% of US jobs at high automation risk (Oxford). But 85% of those will be augmented, not eliminated. Which frame is more honest?"
"The jobs being eliminated first are the entry-level pipeline jobs. What replaces the junior developer → senior developer progression?"
"Which sector's AI displacement is most underreported — legal, pharmacy, or finance? The data might surprise you."
Share this analysis
If this changed how you think about something, share it. The AI workforce conversation needs more data and less hype.
References
All sources accessed March 2026
Live data sources: BLS 2024, Oxford Frey & Osborne, Goldman Sachs AI Research, McKinsey Global Institute, WEF Future of Jobs 2025, OECD Employment Outlook, PwC AI Jobs Barometer