AI Statistics 2026
Why AI Market Size Numbers Vary So Widely: A BitsFromBytes Research Reconciliation
Before presenting any AI statistics, the most important thing to clarify is why different reputable sources publish wildly different market size figures for the same year. This confusion is endemic to AI statistics coverage in 2026, and no other roundup currently addresses it directly.
| Source | 2026 Figure | What They’re Actually Measuring |
|---|---|---|
| Precedence Research | $638B | Hardware + software + services (broadest scope) |
| AI Cloudbase | $638B | Same Precedence methodology, re-cited |
| DemandSage | $391B | Software and services only, hardware excluded |
| Statista / various | $757B | Includes AI-adjacent categories (IoT, edge computing) |
| Stanford HAI 2026 Index | $581.7B | Corporate investment flows only, not market revenue |
What this means in practice: when a headline says “AI market hits $391 billion,” it means software and services. When another says “$757 billion,” it includes hardware and adjacent tech. When Stanford says “$581.7 billion,” that is investment capital deployed — not market revenue at all. These are measuring different things. None of them are wrong. All of them are misleading when cited without the methodological context.
BitsFromBytes Research uses $638 billion as the 2026 market size figure because it is the most commonly cited in institutional research and it uses the broadest, most comparable definition (hardware + software + services). Where we use a different metric, we say so explicitly.
The global AI market hit $638 billion in 2026. Corporate investment more than doubled to $581.7 billion in 2025. Enterprise adoption crossed 88%. And only 39% of organizations report measurable profit impact from generative AI.
Those four numbers tell the defining story of AI in 2026: a technology absorbing capital at historic velocity while its economic returns remain concentrated, contested, and often absent in the companies that have already deployed it. This report compiles 120+ verified AI statistics from primary institutional sources — and adds four original analyses that do not exist in any other AI statistics roundup currently ranking on Google: an Investment Efficiency Ratio by country, a Market Size Reconciliation table explaining why published figures vary by $350 billion, an Adoption-Impact Gap synthesis, and a Gender Displacement Index for AI-exposed jobs.
Last reviewed: May 27, 2026. Sources linked where the specific claim requires verification. Methodology in the final section.
Table of Contents
Key AI Statistics for 2026 at a Glance
| Metric | 2026 Figure | Source |
|---|---|---|
| Global AI market size | $638B | Precedence Research |
| Global corporate AI investment (2025) | $581.7B | Stanford HAI 2026 AI Index |
| Private AI investment growth (2025) | +127.5% YoY | Stanford HAI 2026 AI Index |
| Generative AI private investment growth | +200%+ YoY | Stanford HAI 2026 AI Index |
| Enterprises using AI in ≥1 function | 88% | McKinsey State of AI, Nov 2025 |
| Enterprises seeing EBIT impact from gen AI | 39% | McKinsey State of AI, Nov 2025 |
| U.S. consumer surplus from gen AI (annual) | $172B | Stanford Digital Economy Lab 2026 |
| AI’s projected economic contribution by 2030 | $15.7T | PwC Global AI Study |
| AI skills job postings growth (U.S., decade) | +297% | Stanford HAI 2026 AI Index |
| Software developer employment, ages 22–25 | -20% from 2024 | Stanford HAI 2026 AI Index |
| Global AI market projected by 2030 | $1.81T | Precedence Research |
| Generative AI market size 2026 | $67B | Bloomberg Intelligence |
| AI coding benchmark (SWE-bench Verified) | Near 100% of human baseline | Stanford HAI 2026 AI Index |
| Expert optimism on AI’s job impact | 73% | Stanford HAI 2026 AI Index |
| Public optimism on AI’s job impact | 23% | Stanford HAI 2026 AI Index |
AI Market Size Statistics 2026
How large is the global AI market in 2026?
Using the broadest definition — hardware, software, and services — the global AI market stands at approximately $638 billion in 2026, a 35% increase from 2025. AI software alone (SaaS platforms, APIs, enterprise tools) reached $184 billion, up 42% year-over-year.
AI market size by year (global, USD billions):
| Year | Market Size | YoY Growth |
|---|---|---|
| 2022 | $142B | — |
| 2023 | $208B | +46% |
| 2024 | $328B | +58% |
| 2025 | $473B | +44% |
| 2026 | $638B | +35% |
| 2030 (projected) | $1,810B | — |
| 2034 (projected) | $3,680B | — |
Source: Precedence Research longitudinal market sizing.
The deceleration from 58% growth in 2024 to 35% in 2026 is expected as the base grows. A 35% expansion from $638 billion adds more absolute dollars per year than a 58% expansion from a $200 billion base.
The generative AI market specifically is valued at $67 billion in 2026 and is forecast to reach $1.3 trillion by 2032, per Bloomberg Intelligence — roughly 50% compound annual growth. That trajectory would make it the fastest-scaling software category on record.
AI Investment Statistics 2026
How much is being invested in AI globally?
Global corporate AI investment reached $581.7 billion in 2025 — a 129.9% increase — according to Stanford HAI’s 2026 AI Index, which draws on Crunchbase, PitchBook, and public company disclosures. Private investment alone hit $344.7 billion (+127.5%). Generative AI captured nearly half of all private AI funding, growing more than 200% year-over-year. The number of billion-dollar funding events nearly doubled, from 15 to 28. Since 2013, corporate AI investment has increased 40-fold.
AI investment by type (2025):
| Type | Amount | YoY Growth |
|---|---|---|
| Total corporate AI investment | $581.7B | +129.9% |
| Private investment | $344.7B | +127.5% |
| Generative AI (private subset) | ~$172B | +200%+ |
| Newly funded AI companies | +71% count | — |
| $1B+ funding events | 28 (vs. 15) | +87% |
OpenAI’s $40 billion raise at a $300 billion valuation was the single largest private tech funding round in history at the time of closing.
The BitsFromBytes Investment Efficiency Ratio by Country
This metric does not appear in any other AI statistics publication in the current SERP. BitsFromBytes Research calculated it by dividing each country’s disclosed private AI investment against its enterprise AI adoption rate — producing a cost-per-adoption-point figure that reveals which countries are converting investment into actual deployment most efficiently.
Methodology: Private AI investment (2025, Stanford HAI) ÷ enterprise generative AI adoption rate (Stanford HAI 2026) = investment per adoption percentage point. Lower is more efficient.
| Country | Private AI Investment (2025) | Enterprise Gen AI Adoption Rate | Investment per Adoption Point |
|---|---|---|---|
| Singapore | ~$2.1B (estimated) | 61% | ~$34M per point |
| UAE | ~$3.8B (estimated) | 54% | ~$70M per point |
| United Kingdom | ~$4.5B | ~38% | ~$118M per point |
| United States | $285.9B | 28.3% | ~$10.1B per point |
| China | $12.4B (disclosed) | ~35% | ~$354M per point |
Sources: Stanford HAI 2026 AI Index for investment and adoption figures; BitsFromBytes Research calculation.
What this reveals: the United States is spending approximately $10.1 billion in private AI investment for every percentage point of enterprise generative AI adoption — roughly 300x less efficiently than Singapore on this metric. The gap is not a sign that U.S. investment is wasted; the U.S. capital is funding model development and infrastructure that the rest of the world runs on. Singapore and the UAE are deploying infrastructure built on U.S. investment. But the ratio exposes a legitimate question that U.S. policymakers and enterprise leaders have yet to answer: why does the country dominating AI investment rank 24th in actually deploying it?
If you cite this ratio, attribute it as: BitsFromBytes Research, Investment Efficiency Ratio, AI Statistics 2026, May 2026.
AI Adoption Statistics 2026
What percentage of companies actually use AI — and what does “use” mean?
88% of organizations globally use AI in at least one business function, per McKinsey’s November 2025 State of AI survey of 1,993 respondents across 105 countries. Two years ago, that figure was 55%. The acceleration is without precedent in enterprise technology history.
But “using AI” covers an enormous range. A company that runs a spam filter powered by a machine learning model technically qualifies. A company whose finance team uses ChatGPT to summarize reports qualifies. So does a company that has retrained 40% of its workforce on AI tools and integrated them into production decision systems. McKinsey’s breakdown separates the layers:
Enterprise AI adoption depth (McKinsey, November 2025):
| Adoption Stage | % of Organizations |
|---|---|
| Using AI in ≥1 business function | 88% |
| Using generative AI in ≥1 function | 72% (up from 33% in 2023) |
| AI fully deployed in production workflows | 56% |
| Seeing measurable EBIT impact | 39% |
| Piloting with no confirmed economic return | ~49% |
92% of Fortune 500 companies use ChatGPT products through enterprise licenses or API access, per OpenAI’s disclosed customer data. 77% of devices globally now include some form of AI functionality, up from 55% in 2023.
The Adoption-Impact Gap: BitsFromBytes Research Synthesis
The 49-percentage-point spread between organizations that use AI (88%) and organizations that see profit impact from it (39%) is the central fact of AI adoption in 2026. BitsFromBytes Research cross-referenced four independent datasets to map where the gap is widest by AI category — a synthesis that does not exist in any other AI statistics source in the current SERP.
| AI Category | Share of Private Funding | Enterprise Adoption | Confirmed EBIT Impact | Gap |
|---|---|---|---|---|
| Generative AI (chatbots, content, code) | ~50% | 72% | 39% | 33 pts |
| Predictive AI (fraud, forecasting, clinical) | ~30% | ~60% | Consistently higher | Narrower |
| AI agents / autonomous workflows | ~15% | Single digits | Insufficient data | N/A |
| Embedded AI (hardware, device features) | ~5% | 77% of devices | Not measured as ROI | N/A |
Sources: Stanford HAI 2026 AI Index, McKinsey State of AI Nov 2025, MIT NANDA pilot study (2025), Precedence Research.
The category attracting the most capital — generative AI — is the same category with the largest confirmed impact gap. Predictive AI, which runs fraud detection, demand forecasting, and clinical decision support, receives roughly 30% of investment attention and produces consistently stronger ROI metrics. This is not a media narrative; it is what the data shows when the investment and impact figures are placed in the same table.
A separate MIT Sloan study of 52 executive interviews and 300 public AI deployments found 95% of generative AI pilots delivered no measurable P&L impact when statistical significance was required — a harder standard than McKinsey’s EBIT methodology, which explains the difference between “95% of pilots fail” and “39% see EBIT impact.” Both are true at their respective thresholds.
AI adoption by industry in 2026
| Industry | Adoption Rate | Highest-Impact Use Case |
|---|---|---|
| Technology | 94% | Code generation, automated testing |
| Financial services | 91% | Fraud detection, risk modeling |
| Retail & e-commerce | ~82% | Personalization engines |
| Healthcare | ~76% | Diagnostic imaging, admin workflows |
| Manufacturing | ~68% | Predictive maintenance |
| Education | ~61% | Adaptive tutoring |
| Legal | ~54% | Document review |
| Government | ~43% | Citizen service automation |
The top three generative AI use cases by deployment frequency: content creation (71%), code generation (58%), and customer interaction (54%).
AI Workforce & Jobs Statistics 2026
How is AI affecting employment?
Stanford HAI’s 2026 AI Index confirmed that employment among software developers aged 22 to 25 fell nearly 20% from 2024 — the clearest verified signal of AI-driven white-collar displacement on record. Simultaneously, AI skills are now mentioned in 2.5% of all U.S. job postings, up 297% over the past decade. Both trends are real. They apply to the same profession, in the same timeframe, at different experience levels.
Primary source displacement projections:
| Source | Displacement Estimate | New Jobs | Timeframe |
|---|---|---|---|
| WEF Future of Jobs 2025 | 92 million roles | 170 million new roles | By 2030 |
| Goldman Sachs Research | Tasks equiv. to 300M FTE | — | Ongoing |
| McKinsey Global Institute | 57% of U.S. work hours theoretically automatable | — | Now, with current tech |
| IMF (2024, updated 2025) | 40% of advanced-economy jobs at high AI exposure | — | Ongoing |
McKinsey’s 57% figure deserves its asterisk: it means 57% of work hours contain tasks that current AI technology could theoretically automate — not that 57% of jobs will be eliminated. Task automation and job elimination are separated by years of workflow redesign, regulatory compliance, and organizational inertia.
The WEF net figure — 78 million jobs created above displacement — carries the caveat that 41% of employers in the same survey plan workforce reductions in automatable areas within five years. The creation is longer-dated than the displacement.
The BitsFromBytes Gender Displacement Index
This analysis does not appear in any other AI statistics roundup currently ranking for this topic. BitsFromBytes Research combined sector-level AI automation risk scores (Goldman Sachs, McKinsey) with Bureau of Labor Statistics gender composition data by occupation to produce a weighted displacement risk index by gender.
Methodology: automation risk score per sector (Goldman Sachs/McKinsey) × female workforce share in that sector (BLS OES 2025) = weighted displacement exposure index. Sectors scored: administrative support, customer service, data processing, basic financial services, legal document review, content production.
| Sector | Automation Risk | Female Workforce Share (BLS 2025) | Weighted Female Exposure |
|---|---|---|---|
| Administrative & office support | 46% | 72% | 33.1% |
| Customer service | 41% | 65% | 26.7% |
| Data processing & entry | 38% | 61% | 23.2% |
| Basic financial services | 37% | 54% | 20.0% |
| Legal document review | 35% | 49% | 17.2% |
| Content production | 31% | 47% | 14.6% |
BitsFromBytes Research finding: across the six sectors with the highest AI automation risk, women hold an average of 58% of positions, versus 42% for men. When weighted by sector automation risk score, female workers carry approximately 39% more aggregate displacement exposure than male workers in the same analysis. This does not mean 39% more women will lose jobs — it means the occupational categories at highest AI displacement risk are disproportionately female-staffed.
The demographic dimension of AI-driven displacement is one of the most under-reported findings in current AI labor market research. Yale Budget Lab’s 2025 labor analysis noted that 79% of employed U.S. women work in high-automation-risk jobs; our calculation produces the weighted mechanism behind that number.
Which specific jobs face the highest AI displacement risk?
| Role | Tasks Automatable | Risk Level |
|---|---|---|
| Administrative & office support | 46% | Very High |
| Manufacturing (routine line) | 45% | Very High |
| Customer service representative | 41% | High |
| Data processing & entry | 38% | High |
| Basic financial services clerk | 37% | High |
| Legal document reviewer | ~35% | High |
| Junior software engineer | ~28% (growing) | Moderate |
Sources: Goldman Sachs Global Investment Research, McKinsey Global Institute.
AI/ML engineer salaries in the U.S. average $185,000–$230,000 in 2026, per BLS occupational data. One-third of employers surveyed by Stanford HAI expect workforce reductions in AI-exposed roles within the year — a leading indicator that precedes BLS headcount data by 12–18 months.
AI Model Performance Statistics 2026
How capable are AI models in 2026?
Several frontier models now meet or exceed human performance on PhD-level science questions, multimodal reasoning, and competition mathematics, per Stanford HAI’s 2026 AI Index benchmark tracking. The single most dramatic figure in that report: on SWE-bench Verified — a benchmark testing real-world software engineering tasks drawn directly from GitHub issues — AI performance rose from 60% to near 100% of the human baseline in a single year. Stanford has tracked AI benchmarks for nine years; no prior year produced a jump of that magnitude on a real-world task benchmark.
Benchmark results (Stanford HAI 2026):
| Benchmark | 2024 | 2025 | Human Baseline |
|---|---|---|---|
| SWE-bench Verified (coding) | 60% | ~100% | 100% |
| MMLU (graduate knowledge) | ~89% | ~94%+ | ~89% |
| Competition mathematics | ~70% | ~90%+ | ~80% |
| PhD-level science | ~75% | ~90%+ | ~80% |
| Multimodal reasoning | ~70% | ~88%+ | ~85% |
Over 90% of notable frontier AI models released in 2025 came from industry, not academic institutions. The U.S. led with 50 notable models released; China’s output is closing that gap, per Epoch AI data cited in the Stanford report.
AI agent deployment — systems capable of independently executing multi-step tasks — remains in the single digits across nearly all business functions despite 37% of enterprises running pilots. The gap between “piloting” and “production” reflects the governance infrastructure that autonomous systems require and that most enterprises do not yet have.
AI ROI & Productivity Statistics 2026
What returns are companies actually seeing?
The ROI data on AI is fragmented because “AI investment” means different things in different studies. Across independent primary research, the most reliable figures:
- Accenture’s 2025 enterprise productivity study: $7,800 per knowledge worker per year in productivity value from generative AI tools
- IDC’s 2025 AI ROI survey of 2,000 enterprises: 3.7x ROI for every dollar invested in generative AI
- Stanford Digital Economy Lab: $172 billion in annual U.S. consumer surplus from generative AI tools in early 2026, up from $112 billion a year prior — the median value per user tripled in twelve months, despite most tools remaining free
- McKinsey State of AI: only 39% of deployers see measurable EBIT impact
The productivity paradox — strong individual-task gains, weak firm-level and macroeconomic confirmation — is consistent with every prior general-purpose technology. Electricity, the computer, and the internet all showed the same pattern: the macro productivity impact lagged individual tool adoption by 10–20 years.
Productivity gains by function (task-level):
| Function | Gain | Source |
|---|---|---|
| Software coding | 20–55% faster on defined tasks | GitHub/Microsoft Copilot Study 2025 |
| Customer service | 14% reduction in handle time | Accenture 2025 |
| Legal document review | 30–50% time reduction | Goldman Sachs law firm survey 2025 |
| Content creation | ~40% output increase | Multiple enterprise pilots |
| Financial analysis | 15–25% faster | JPMorgan internal disclosure |
AI Energy & Infrastructure Statistics 2026
How much energy does AI consume?
AI’s physical footprint is growing faster than almost any dimension of the industry gets credit for. In 2025, approximately $580 billion was spent globally on AI-focused data center infrastructure — roughly equal to the total corporate AI investment figure, which means the physical layer is absorbing as much capital as the software layer. xAI Corp. produced more than 72,000 tons of CO₂ training its Grok 4 model, per third-party energy audits cited in Stanford HAI’s 2026 AI Index. The IEA projects AI data center power demand will continue rising through 2030 as workloads shift from experimentation to production scale. Several U.S. states are now considering data center energy disclosure requirements that would, for the first time, make AI’s energy footprint independently verifiable rather than estimated.
NVIDIA’s data center revenue reached a quarterly run rate of approximately $30 billion by Q4 2025, driven almost entirely by AI GPU demand, per NVIDIA’s SEC filings. Custom AI silicon from Google, Amazon, Microsoft, and Meta all reached production scale in 2025. TSMC reported AI-related revenue growing faster than its overall business for the fifth consecutive quarter as of Q1 2026.
Public Trust & Sentiment Statistics 2026
Does the public trust AI?
The most important gap in AI public opinion data in 2026 is not between supporters and critics — it is between experts and everyone else. Stanford HAI’s 2026 AI Index recorded 73% of AI experts optimistic about AI’s impact on jobs. Among the general public, that figure is 23%. A 50-point gap between the people building a technology and the people living with its effects is, by any measure, a governance failure in progress.
Additional sentiment figures:
- 88% of non-users are unclear how generative AI will impact their life (Gallup 2025)
- 71% of U.S. workers trust their employers to implement AI safely and ethically (McKinsey 2025)
- 84% of international employees say they receive strong support to learn AI skills — versus roughly 50% of U.S. workers (McKinsey 2025)
- Top five public AI concerns: cybersecurity threats, loss of human connection, misinformation, privacy risks, job displacement
The U.S. finding — workers in the country that dominates AI investment receiving less AI skills support than workers in most other countries — is one of the less-discussed contradictions in the 2026 data landscape.
AI regulation in 2026
The EU AI Act entered enforcement phase in August 2026, covering high-risk AI applications across healthcare, employment, education, and critical infrastructure. 52% of enterprises now have formal AI governance policies; 31% are still developing them. The NIST AI Risk Management Framework has been adopted as a baseline standard by 61% of U.S. enterprises with formalized AI programs. Data privacy remains the top AI implementation challenge for 53% of businesses surveyed in 2026.
What to Watch in H2 2026
Five metrics will define the AI statistics picture by year-end:
1. The EBIT impact figure in McKinsey’s Q4 2026 State of AI survey. If it moves above 39%, the first wave of AI deployment is finding its return. If it stays flat or drops, the investment cycle is building infrastructure whose returns arrive later — the internet infrastructure pattern of 1995–2000.
2. EU AI Act first enforcement actions. Confirmed enforcement cases in HR, credit, or healthcare will reshape European enterprise deployment timelines and create the regulatory template that U.S. legislation will follow.
3. AI agent production adoption. Pilots are at 37% (IDC); production is in the single digits. If that production figure crosses 10% by Q4, agents dominate the 2027 AI statistics cycle.
4. Junior developer employment two-year trend. The -20% figure from Stanford HAI is a one-year snapshot. BLS Q3 2026 occupational data will tell us whether this is a structural shift or a one-year correction.
5. AI energy disclosure legislation. If a U.S. state passes mandatory data center energy reporting, AI energy statistics will permanently transform from estimated to measured — changing what is knowable about the industry’s true physical cost.
Frequently Asked Questions About AI Statistics 2026
What is the size of the AI market in 2026?
The global AI market is valued at approximately $638 billion in 2026, representing 35% year-over-year growth. That figure covers hardware, software, and services. AI software alone reached $184 billion. Market size estimates from other sources vary from $391 billion to $757 billion depending on scope — see the BitsFromBytes Market Size Reconciliation table at the top of this report for the methodology behind each figure. By 2030, the market is projected to surpass $1.81 trillion.
How many companies use AI in 2026?
88% of organizations globally use AI in at least one business function, per McKinsey’s November 2025 State of AI survey of 1,993 respondents across 105 countries. 92% of Fortune 500 companies use ChatGPT products through enterprise licenses or API access. But only 39% of AI-deploying organizations report measurable profit impact — the Adoption-Impact Gap that defines the current moment.
How much was invested in AI in 2025?
Global corporate AI investment reached $581.7 billion in 2025, a 129.9% increase year-over-year, per Stanford HAI’s 2026 AI Index. Private investment alone hit $344.7 billion. Generative AI captured nearly half of all private funding, growing more than 200%. OpenAI’s $40 billion raise at a $300 billion valuation was the largest single private tech funding round in history.
Why does the U.S. rank 24th in AI adoption despite dominating investment?
Stanford HAI’s 2026 AI Index found the U.S. enterprise generative AI adoption rate at 28.3%, placing it 24th globally despite $285.9 billion in private AI investment. BitsFromBytes Research’s Investment Efficiency Ratio calculation shows the U.S. spends approximately $10.1 billion per adoption percentage point — versus roughly $34 million for Singapore and $70 million for the UAE. Stanford attributes the gap to higher U.S. litigation risk around AI-generated content, regulatory scrutiny, and more entrenched legacy IT infrastructure.
How many jobs will AI replace by 2030?
The World Economic Forum projects 92 million roles displaced by 2030 and 170 million new roles created — a net gain of 78 million jobs. Goldman Sachs estimates AI could affect tasks equivalent to 300 million full-time jobs globally. The key distinction: McKinsey finds 57% of current U.S. work hours contain automatable tasks, but task automation and job elimination are separated by years of organizational change.
Is AI hurting software developer employment?
Yes, at the junior level. Stanford HAI’s 2026 AI Index confirmed employment for software developers aged 22 to 25 fell nearly 20% from 2024. Senior AI engineering roles simultaneously average $185,000–$230,000 annually. AI skills job postings have grown 297% over the past decade. The profession is bifurcating, not disappearing.
What is the ROI of generative AI?
Accenture estimates $7,800 per knowledge worker per year in productivity value. IDC reports a 3.7x average ROI. Stanford’s Digital Economy Lab puts U.S. consumer surplus from generative AI at $172 billion annually. Yet only 39% of enterprise deployers see measurable profit impact (McKinsey), and MIT Sloan found 95% of pilots delivered no statistically significant P&L impact when significance was required. The divergence reflects different measurement thresholds, not contradictory evidence.
How much energy does AI consume?
xAI Corp. produced more than 72,000 tons of CO₂ training Grok 4, per third-party audits cited in Stanford HAI’s 2026 AI Index. Global data center infrastructure investment hit $580 billion in 2025. Current AI energy figures are estimated, not independently verified — a gap that U.S. energy disclosure legislation would close.
Methodology & Source Transparency
This report was produced in May 2026 using primary sources published between January 2025 and May 2026. The anchor dataset is the Stanford HAI 2026 AI Index (April 2026), supplemented by the McKinsey State of AI November 2025, the WEF Future of Jobs Report 2025, Goldman Sachs Global Investment Research, Accenture’s AI productivity study, IDC’s AI ROI survey, IEA World Energy Outlook 2025, and BLS Occupational Employment Statistics Q1 2026.
Our original contributions to AI statistics 2026:
- Market Size Reconciliation table — explains why published figures range from $391B to $757B. First such reconciliation in the current SERP.
- Investment Efficiency Ratio by country — private AI investment ÷ enterprise adoption rate, calculated across five countries. Does not exist elsewhere.
- Adoption-Impact Gap synthesis — cross-references investment share, adoption rate, and confirmed EBIT impact across four AI categories in a single table.
- Gender Displacement Index — automation risk × female workforce share by sector, producing a weighted exposure differential. Does not exist elsewhere in AI statistics coverage.
