This hub gathers evidence on how AI affects work, the economy, the environment, and everyday life in Canada and the United States. It is written for readers who want official data, named studies, and specific rules instead of forecasts, and it links to the detailed article on each question.
Key takeaways
- Data centers used about 1.5 percent of world electricity in 2024, according to the International Energy Agency (IEA).
- The US Bureau of Labor Statistics (BLS) projects total job growth of 3.5 percent from 2025 to 2035.
- Official labor releases in Canada and the US do not record a cause for unemployment, so AI's role stays open.
- Exposure to AI is an upper limit on what could change, not a count of jobs lost.
- Rules in Ontario, New York City, California, and Quebec already set requirements for AI in hiring and automated decisions.
- Jobs and labor data
- Occupations, oversight, and working with AI
- Economy, funding, and how AI is built and governed
- Environment, energy, farming, and water
- Accessibility and schools
- FAQ
Jobs and labor data
These articles compare what official surveys, projections, and layoff trackers measure about AI and employment. They separate exposure to AI from displacement, and show where agencies do and do not name AI as a cause.
- Does AI Create Jobs? Three Kinds of Evidence Compared
- AI and Unemployment: What 2026 Labor Data Actually Show
- Is AI Cutting Entry-Level Jobs? Canadian and US Evidence
- How Official Job Projections Account for AI, 2025 to 2035
- Will AI Replace Most Jobs? What ILO, IMF, and Yale Data Show
- Job Displacement Data: What Official Surveys Show About AI
- Augmentation or Automation? How the ILO Sorts Jobs by AI Exposure
- AI and Job Loss: Layoff Announcements and Projections, 2026
- Will AI Take White-Collar Jobs? What 2026 Entry-Level Data Show
Occupations, oversight, and working with AI
Individual professions and workplaces raise narrower questions. These articles cover scientists, accountants, and nurses, the rules on hiring and human review of automated decisions, and research on how people and AI perform together.
- Can AI Replace Scientists? What Labor Data and Studies Show
- Will AI Replace Accountants? What Canada's Job Bank Shows
- Will AI Replace Nurses? What Workforce Data Shows
- AI in Hiring: What Ontario and New York City Require
- Will AI Fully Replace Humans? Where Rules Require a Person
- AI Employers: How to Judge Workplace Quality With Evidence
- Humans Plus AI: What 106 Experiments Found on Teamwork
- Human-AI Interaction: The 18 CHI Guidelines and What Tests Found
- AI and Handymen: What Repair-Work Data Show About Chatbots
Economy, funding, and how AI is built and governed
Official productivity data and economic models give different pictures of AI's effect on growth. This group also covers grant programs, how a large language model is built, open source definitions, and the bodies that test or coordinate AI internationally.
- AI and Productivity: What US and Canadian Official Data Show
- AI and Economic Growth: What Estimates and Data Show
- AI Grants: Real Funding Programs in Canada and the US
- How a Large Language Model Is Built: Inside the Llama 3 Report
- Water Jug Problem in AI: State-Space Search With Python Code
- Open Source AI: What the Definition Requires and Allows
- How Governments Test OpenAI Models: What Is Documented
- Global Partnership on AI: Members, Structure, and Role
- AI Investment in US GDP Growth: Fed Estimates, 2025 to 2026
- AI Reshaping Industries? One Treasury Case, Checked
Environment, energy, farming, and water
AI uses electricity and is also applied to environmental problems. These articles cover energy demand and measured benefits, satellite monitoring of fishing, solar forecasting, crop mapping, farm results, and water and sanitation.
- AI and the Environment: Energy Use and Measured Benefits
- AI for the Environment: Spotting Untracked Ships From Space
- AI Solar Forecasting: What a Better PV Forecast Is Worth
- AI in Agriculture: Measured Results and Canadian Adoption
- AI Crop Maps From Space: How Accurate Are They?
- AI in Water and Sanitation: What Is Documented and What Is Not
Accessibility and schools
Assistive AI builds on the independent living movement and on accessibility rules for schools. These articles cover speech recognition research, school AI tools, and who decides about AI in school districts.
- Ed Roberts and the Independent Living Movement, Explained
- AI Tools for Students With Disabilities: Accessibility Rules
- Speech Recognition for Disabilities: What One Project Found
- School Districts and AI: Who Decides in the US and Ontario
- AI in Education in 2023: What Was Said, What Holds in 2026
FAQ
How much electricity do data centers use?
The IEA estimates that data centers used about 415 terawatt-hours in 2024, roughly 1.5 percent of world electricity. Its base case projects about 945 terawatt-hours by 2030, just under 3 percent.
Do official projections expect AI to cut total employment?
The BLS projects total employment to grow about 3.5 percent from 2025 to 2035. Its projections assume AI follows historical patterns, and office support is the largest projected decline.
Does AI exposure mean a job will disappear?
No. The ILO's 2025 index treats exposure as an upper limit on possible change, not as a count of jobs lost. One in four workers have some generative AI exposure, and 3.3 percent are in the highest tier.
Do unemployment figures show that AI is causing job losses?
Not directly. The official September 2026 releases for Canada (6.5 percent) and the US (4.2 percent) record no cause, so AI's role remains open.
Related reading
More in AI Blog.
Last verified: October 2026
Sources
- Energy and AI, Energy demand from AI, International Energy Agency, 2025
- Employment by major occupational group, 2025 and projected 2035, US Bureau of Labor Statistics, 2026
- Generative AI and Jobs: A Refined Global Index of Occupational Exposure, International Labor Organization, 2025