## 1. AI Preparedness and Investment

## Source details

**Canonical URL:** [1. AI Preparedness and Investment](https://www.imf.org/-/media/files/publications/selected-issues-papers/2026/english/sipea2026061.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/selected-issues-papers/2026/english/sipea2026061.pdf.md)
- [Structured JSON version](/-/media/files/publications/selected-issues-papers/2026/english/sipea2026061.pdf.json)

---

### A. Context
- Generative AI expands AI scope beyond conventional machine-learning applications to tasks such as pattern recognition, content creation, and decision support, opening the door to sizable productivity gains.
- AI has diffused relatively quickly across firms and occupations (Mish et al., 2025).
- This chapter maps AI exposure and complementarity onto occupational microdata for Israel and benchmarks against selected European economies.

### B. Preparedness and Challenges
- Israel’s position and adoption:
  - IMF’s AI Preparedness Index ranks Israel 18th globally.
  - Across all sectors, about 28 percent of Israeli firms report using AI.
  - Israel is among leading OECD countries in current and projected business-sector AI adoption.
- Constraints and imbalances:
  - The IMF’s Skill Readiness and Imbalance Index indicates demand significantly exceeds supply; values are normalized between –1 and 1 with the US set to zero as benchmark.
  - The gap in graduates with new skills is larger in Israel than in many advanced economies, reflecting a relatively low number of graduates in AI-relevant fields.
  - Israel’s “dual economy”: high skills and productivity in high-tech versus weaker skills and lagging digitization in other sectors; AI use concentrated in high-tech and ICT.
  - Significant skill gaps persist across population groups, especially among Arab-Israeli and Haredi populations:
    - Haredi men: limited participation in mathematics, science, and English.
    - Underrepresentation in high-tech employment reflected by an index where values below one indicate underrepresentation.
- AI Preparedness Index by factors (Israel = 0.73) highlights strengths across:
  - Regulation and Ethics; Human Capital and Labor Market Policies; Innovation and Economic Integration; Digital Infrastructure.

### C. Labor Market Exposure and Complementarity to AI
- Methodology and classification:
  - Builds on Felten et al. (2021) and Pizzinelli et al. (2023), grouping occupations into HEHC (high exposure, high complementarity), HELC (high exposure, low complementarity), and LE (low exposure).
  - Analysis uses 2023 Israel, Eurostat and U.K. Labor Force Surveys at ISCO two-digit level, covering 36 occupations, 19 economic activities, three age groups, gender, three educational levels, and three population groups.
- Key results and distributions:
  - About two-thirds of Israeli workers face high AI exposure—broadly in line with the U.K., but higher than other European countries.
  - Employment shares by exposure/complementarity (Israel):
    - High Exposure, High Complementarity (HEHC): 44.1 percent
    - High Exposure, Low Complementarity (HELC): 23.1 percent
    - Low Exposure (LE): 32.8 percent
  - Among highly exposed workers, roughly two-thirds are in complementary occupations (education; business and administration associate professionals; science and engineering; health professions).
  - The remaining third of highly exposed workers exhibits lower complementarity, including sizable employment shares in ICT professionals and sales occupations, implying higher displacement risks.
  - About one-third of the workforce has low AI exposure, concentrated in personal care, personal services, and manual labor.
  - Sectoral and demographic drivers of Israel’s higher HEHC share include a younger population (increasing employment in education), a large high-tech sector, and a higher share of legal, social, and cultural professionals.
- Private vs Public sector:
  - In the private sector, about one-third of employment faces high displacement risks (HELC concentrated in business and administration professionals, sales workers, ICT specialists).
  - ICT, finance, and insurance together account for about 12 percent of total employment.
  - Public services account for about one-third of total employment, with more than half of public sector workers in HEHC occupations (education and healthcare), suggesting the public sector is likely to benefit from AI.
- Comparative findings:
  - Compared with peers, displacement risks in Israel are lower than in the United Kingdom and other European economies, but realizing benefits depends on effective AI adoption.

### D. Sectoral Differences
- Exposure and complementarity vary across sectors:
  - Broad-sector exposure/complementarity (selected high-level observations):
    - Education and Human health and social work activities show large HEHC shares.
    - Information and communication and Financial and insurance activities show higher HELC shares.
  - Employment by sector (percentage of total employment, 2023) differs across Israel, UK, benchmark European, and other European samples, influencing aggregate exposure patterns.
- Main occupations in Israel (top 20 occupations by percentage of total employment, 2023) include:
  - Teaching Professionals
  - Business and Administration Associate Professionals
  - Personal Care Workers
  - Information and Communications Technology Professionals
  - Legal/Social/Cultural Professionals
  - Sales Workers
  - Science and Engineering Professionals
  - Personal Service Workers
  - Business and Administration Professionals
  - (and others listed among the top 20)

### E. Gender, Education, and Age Differences
- Gender differences:
  - Exposure/complementarity for women (Israel):
    - HEHC: 50.6 percent
    - HELC: 42.2 percent
    - LE: 7.2 percent (implied by chart structure)
  - Exposure/complementarity for men (Israel):
    - HEHC: 37.7 percent
    - HELC: 37.3 percent
    - LE: 25.0 percent (implied by chart structure)
  - Nearly three-quarters of female workers are highly exposed to AI, with most in HEHC occupations (education, business and administration associate professions, legal/social/cultural services, healthcare). Israeli women are likely to gain more than women in Europe due to concentration in education and healthcare.
- Population group differences (employment exposure and complementarity to AI, percent; population shares in parentheses):
  - non-Haredi Jews (74%): HEHC 48.22; HELC 25.84; LE 25.94
  - Haredi Jews (9%): HEHC 55.13; HELC 20.54; LE 24.33
  - Arab-Israeli (17%): HEHC 27.47; HELC 14.86; LE 57.67
  - Around 25 percent of non-Haredi Jewish workers and 20 percent of Haredi workers are employed in HELC occupations. Arab-Israelis are less exposed overall.
- Education and age:
  - Education (employment exposure and complementarity, percent):
    - Less than primary, primary and lower secondary education: HEHC 7.7; HELC 31.3; LE 66.4
    - Upper secondary and post-secondary non-tertiary education: HEHC 11.1; HELC 23.2; LE 25.3
    - Tertiary education: HEHC 81.1; HELC 45.5; LE 8.3
  - Age (employment exposure and complementarity, percent):
    - Less than 30: HEHC 31.7; HELC 48.5; LE 19.8
    - 30-45: HEHC 46.7; HELC 28.7; LE 24.6
    - More than 45: HEHC 34.6; HELC 27.2; LE 38.2
  - Highly educated workers: nearly 90 percent are exposed to AI, and about one-quarter may experience displacement pressures.
  - Younger workers appear more vulnerable than older cohorts because higher exposure is not consistently offset by complementarity.

### F. Conclusions and Policy Considerations
- Summary assessment:
  - AI offers significant potential to raise productivity across traditional industries and the public sector, but poses adjustment challenges.
  - Rapid technological change has increased demand for skilled workers beyond supply.
  - Israel’s labor market is less susceptible to AI-related job displacement than those of other European economies due to higher complementarity, yet around one-fifth of workers remain at risk.
  - Vulnerable groups concentrate in high-value added industries (ICT, finance, professional services), younger workers, and those with tertiary education.
- Policy recommendations:
  - Align workforce skills with evolving needs:
    - Education reforms to expand STEM enrollment, update mathematics and science curricula, and introduce AI education at earlier stages.
    - Develop a comprehensive lifelong learning strategy—reskilling, upskilling, and mid-career training.
  - Strengthen active labor market policies (ALMPs): Israel’s ALMP expenditure (Percent of GDP) is lower than peers; chart shows cross-country values for 2017–2021 with Israel among the lower end.
  - Targeted inclusion policies for Arab-Israeli and Haredi populations:
    - Continue and scale past initiatives: MAHAT technological training programs; gap-year transition programs for Arab-Israeli youth; employment support through Ryan Centers; Haredi-focused coding and placement initiatives.
    - These initiatives align with the 2022 Perlmutter Committee recommendations (upgrade seminary training; better align vocational education and training with labor market needs).

### Annex I. Measuring Exposure and Complementarity to AI
- 1. Assessing Occupational Exposure to AI.
  - Felten et al. (2021) introduce an approach to assessing AI exposure across occupations by associating a set of AI capabilities, such as image recognition and text generation, with a broad range of underlying worker abilities identified in the U.S. O*NET database.
  - The framework treats occupations as portfolios of skills, where each ability contributes to job performance with varying degrees of importance and complexity, allowing exposure to AI to be quantified at the occupational level.
- 2. Assessing Complementarity with AI.
  - Pizzinelli et al. (2023) construct an index capturing how AI interacts with different occupations by incorporating two additional dimensions from the O*NET database: work contexts and job zones.
  - The authors identify eleven AI-relevant work contexts and combine them with job-zone classifications to form six broader components—communication, responsibility, physical conditions, criticality, routine intensity, and skills.
  - This framework allows for an evaluation of the extent to which AI technologies are likely to complement or substitute tasks performed within specific occupations.
- 3. Combining Exposure and Complementarity.
  - Exposure and complementarity measures can be jointly interpreted within a three-category framework that classifies occupations according to their degree of AI exposure and complementarity.
  - Using the median values of both indicators as thresholds, occupations are grouped into: high exposure–high complementarity (HEHC), high exposure–low complementarity (HELC), and low exposure (LE).
  - Occupations in the high exposure–low complementarity category are considered more susceptible to AI-driven job displacement, as AI technologies are more likely to substitute for, rather than complement, the tasks they perform.

*Source: IMF staff estimates and analysis in "1. AI Preparedness and Investment" (Israel), Selected Issues Paper, June 10, 2026.*

### 1. AI Preparedness and Investment _____________________________________________________ 3

### 1. AI Preparedness and Investment

### A. Context
- Generative AI expands AI scope beyond conventional machine-learning applications to tasks such as pattern recognition, content creation, and decision support, opening the door to sizable productivity gains.
- AI has diffused relatively quickly across firms and occupations (Mish et al., 2025).
- This chapter maps AI exposure and complementarity onto occupational microdata for Israel and benchmarks against selected European economies.

### B. Preparedness and Challenges
- Israel’s position and adoption:
  - IMF’s AI Preparedness Index ranks Israel 18th globally.
  - Across all sectors, about 28 percent of Israeli firms report using AI.
  - Israel is among leading OECD countries in current and projected business-sector AI adoption.
- Constraints and imbalances:
  - The IMF’s Skill Readiness and Imbalance Index indicates demand significantly exceeds supply; values are normalized between –1 and 1 with the US set to zero as benchmark.
  - The gap in graduates with new skills is larger in Israel than in many advanced economies, reflecting a relatively low number of graduates in AI-relevant fields.
  - Israel’s “dual economy”: high skills and productivity in high-tech versus weaker skills and lagging digitization in other sectors; AI use concentrated in high-tech and ICT.
  - Significant skill gaps persist across population groups, especially among Arab-Israeli and Haredi populations:
    - Haredi men: limited participation in mathematics, science, and English.
    - Underrepresentation in high-tech employment reflected by an index where values below one indicate underrepresentation.
- AI Preparedness Index by factors (Israel = 0.73) highlights strengths across:
  - Regulation and Ethics; Human Capital and Labor Market Policies; Innovation and Economic Integration; Digital Infrastructure.

### C. Labor Market Exposure and Complementarity to AI
- Methodology and classification:
  - Builds on Felten et al. (2021) and Pizzinelli et al. (2023), grouping occupations into HEHC (high exposure, high complementarity), HELC (high exposure, low complementarity), and LE (low exposure).
  - Analysis uses 2023 Israel, Eurostat and U.K. Labor Force Surveys at ISCO two-digit level, covering 36 occupations, 19 economic activities, three age groups, gender, three educational levels, and three population groups.
- Key results and distributions:
  - About two-thirds of Israeli workers face high AI exposure—broadly in line with the U.K., but higher than other European countries.
  - Employment shares by exposure/complementarity (Israel):
    - High Exposure, High Complementarity (HEHC): 44.1 percent
    - High Exposure, Low Complementarity (HELC): 23.1 percent
    - Low Exposure (LE): 32.8 percent
  - Among highly exposed workers, roughly two-thirds are in complementary occupations (education; business and administration associate professionals; science and engineering; health professions).
  - The remaining third of highly exposed workers exhibits lower complementarity, including sizable employment shares in ICT professionals and sales occupations, implying higher displacement risks.
  - About one-third of the workforce has low AI exposure, concentrated in personal care, personal services, and manual labor.
  - Sectoral and demographic drivers of Israel’s higher HEHC share include a younger population (increasing employment in education), a large high-tech sector, and a higher share of legal, social, and cultural professionals.
- Private vs Public sector:
  - In the private sector, about one-third of employment faces high displacement risks (HELC concentrated in business and administration professionals, sales workers, ICT specialists).
  - ICT, finance, and insurance together account for about 12 percent of total employment.
  - Public services account for about one-third of total employment, with more than half of public sector workers in HEHC occupations (education and healthcare), suggesting the public sector is likely to benefit from AI.
- Comparative findings:
  - Compared with peers, displacement risks in Israel are lower than in the United Kingdom and other European economies, but realizing benefits depends on effective AI adoption.

### D. Sectoral Differences
- Exposure and complementarity vary across sectors:
  - Broad-sector exposure/complementarity (selected high-level observations):
    - Education and Human health and social work activities show large HEHC shares.
    - Information and communication and Financial and insurance activities show higher HELC shares.
  - Employment by sector (percentage of total employment, 2023) differs across Israel, UK, benchmark European, and other European samples, influencing aggregate exposure patterns.
- Main occupations in Israel (top 20 occupations by percentage of total employment, 2023) include Teaching Professionals, Business and Administration Associate Professionals, Personal Care Workers, Information and Communications Technology Professionals, Legal/Social/Cultural Professionals, Sales Workers, Science and Engineering Professionals, Personal Service Workers, Business and Administration Professionals, among others.

### E. Gender, Education, and Age Differences
- Gender differences:
  - Exposure/complementarity for women (Israel):
    - HEHC: 50.6 percent
    - HELC: 42.2 percent
    - LE: 7.2 percent (implied by chart structure)
  - Exposure/complementarity for men (Israel):
    - HEHC: 37.7 percent
    - HELC: 37.3 percent
    - LE: 25.0 percent (implied by chart structure)
  - Nearly three-quarters of female workers are highly exposed to AI, with most in HEHC occupations (education, business and administration associate professions, legal/social/cultural services, healthcare). Israeli women are likely to gain more than women in Europe due to concentration in education and healthcare.
- Population group differences (employment exposure and complementarity to AI, percent; population shares in parentheses):
  - non-Haredi Jews (74%): HEHC 48.22; HELC 25.84; LE 25.94
  - Haredi Jews (9%): HEHC 55.13; HELC 20.54; LE 24.33
  - Arab-Israeli (17%): HEHC 27.47; HELC 14.86; LE 57.67
  - Around 25 percent of non-Haredi Jewish workers and 20 percent of Haredi workers are employed in HELC occupations. Arab-Israelis are less exposed overall.
- Education and age:
  - Education (employment exposure and complementarity, percent):
    - Less than primary, primary and lower secondary education: HEHC 7.7; HELC 31.3; LE 66.4
    - Upper secondary and post-secondary non-tertiary education: HEHC 11.1; HELC 23.2; LE 25.3 (values shown in chart)
    - Tertiary education: HEHC 81.1; HELC 45.5; LE 8.3 (chart indicates concentration of HEHC among tertiary but also meaningful HELC share)
  - Age (employment exposure and complementarity, percent):
    - Less than 30: HEHC 31.7; HELC 48.5; LE 19.8 (implied by chart distribution)
    - 30-45: HEHC 46.7; HELC 28.7; LE 24.6
    - More than 45: HEHC 34.6; HELC 27.2; LE 38.2
  - Highly educated workers: nearly 90 percent are exposed to AI, and about one-quarter may experience displacement pressures.
  - Younger workers appear more vulnerable than older cohorts because higher exposure is not consistently offset by complementarity.

### F. Conclusions and Policy Considerations
- Summary assessment:
  - AI offers significant potential to raise productivity across traditional industries and the public sector, but poses adjustment challenges.
  - Rapid technological change has increased demand for skilled workers beyond supply.
  - Israel’s labor market is less susceptible to AI-related job displacement than those of other European economies due to higher complementarity, yet around one-fifth of workers remain at risk.
  - Vulnerable groups concentrate in high-value added industries (ICT, finance, professional services), younger workers, and those with tertiary education.
- Policy recommendations:
  - Align workforce skills with evolving needs:
    - Education reforms to expand STEM enrollment, update mathematics and science curricula, and introduce AI education at earlier stages.
    - Develop a comprehensive lifelong learning strategy—reskilling, upskilling, and mid-career training.
  - Strengthen active labor market policies (ALMPs): Israel’s ALMP expenditure (Percent of GDP) is lower than peers; chart shows cross-country values for 2017–2021 with Israel among the lower end.
  - Targeted inclusion policies for Arab-Israeli and Haredi populations:
    - Continue and scale past initiatives: MAHAT technological training programs; gap-year transition programs for Arab-Israeli youth; employment support through Ryan Centers; Haredi-focused coding and placement initiatives.
    - These initiatives align with the 2022 Perlmutter Committee recommendations (upgrade seminary training; better align vocational education and training with labor market needs).

*Source: IMF staff estimates and analysis in "1. AI Preparedness and Investment" (Israel), Selected Issues Paper, June 10, 2026.*

### References

### sipea2026061 - References

### References
- Bank of Israel, 2025, Annual Report 2024, Box 5.1, “The expected effect of generative Artificial Intelligent on employees: Implications for labor market policy,” March.
- Cazzaniga M., F. Jaumotte, L. Li, G. Melina, A. J. Panton, C. Pizzinelli, E. Rockall, and M. M. Tavares, 2024, “Gen-AI: Artificial Intelligence and the Future of Work”, IMF Staff Discussion Note SDN2024/001, International Monetary Fund, Washington, DC.
- Felten E., M. Raj and R. Seamans, 2021, “Occupational, industry, and geographic exposure to artificial intelligence: A novel dataset and its potential uses”, Strategic Management Journal, 2021;1-23.
- Filippucci, F., Gal, P., Laengle, K., Schief, M. and Yildirim, M.A., 2026. AI meets trade: Global linkages and the cross-country distribution of the gains from AI. OECD Artificial Intelligence Papers No. 57. Paris: OECD Publishing.
- Institute for National Security Studies (INSS), 2025, Recommendations of the Nagel Committee: Rushing Toward the AI Abyss, Memorandum No. 2048.
- International Monetary Fund (IMF). 2026. Bridging Skill Gaps for the Future: New Jobs Creation in the AI Age. Staff Discussion Notes, No. 2026/001. Washington, DC: International Monetary Fund.
- Israel Central Bureau of Statistics (CBS), 2025, “Business Tendency Survey: Artificial Intelligence Module”, June.
- Mish F., B. Park, C. Pizzinelli, and G. Sher., 2025 AI and Productivity in Europe, IMF Working Papers 2025/067, International Monetary Fund, Washington, DC.
- Organisation for Economic Co-operation and Development (OECD), 2025, OECD Economic Surveys: Israel 2025, OECD Publishing, Paris.
- Pizzinelli C., A. Panton, M. M. Tavares, M. Cazzaniga, and L. Li. 2023. "Labor Market Exposure to AI: Cross-Country Differences and Distributional Implications." IMF Working Paper 2023/216, International Monetary Fund, Washington, DC.
- Stanford Institute for Human‑Centered Artificial Intelligence (HAI), 2025, AI Index Report 2025, Stanford University, Stanford, CA.
- State Comptroller of Israel, 2024, “Artificial Intelligence: National Preparedness”, Special Report.

### Annex I. Measuring Exposure and Complementarity to AI
- 1. Assessing Occupational Exposure to AI.
  - Felten et al. (2021) introduce an approach to assessing AI exposure across occupations by associating a set of AI capabilities, such as image recognition and text generation, with a broad range of underlying worker abilities identified in the U.S. O*NET database.
  - The framework treats occupations as portfolios of skills, where each ability contributes to job performance with varying degrees of importance and complexity, allowing exposure to AI to be quantified at the occupational level.
- 2. Assessing Complementarity with AI.
  - Pizzinelli et al. (2023) construct an index capturing how AI interacts with different occupations by incorporating two additional dimensions from the O*NET database: work contexts and job zones.
  - The authors identify eleven AI-relevant work contexts and combine them with job-zone classifications to form six broader components—communication, responsibility, physical conditions, criticality, routine intensity, and skills.
  - This framework allows for an evaluation of the extent to which AI technologies are likely to complement or substitute tasks performed within specific occupations.
- 3. Combining Exposure and Complementarity.
  - Exposure and complementarity measures can be jointly interpreted within a three-category framework that classifies occupations according to their degree of AI exposure and complementarity.
  - Using the median values of both indicators as thresholds, occupations are grouped into: high exposure–high complementarity (HEHC), high exposure–low complementarity (HELC), and low exposure (LE).
  - Occupations in the high exposure–low complementarity category are considered more susceptible to AI-driven job displacement, as AI technologies are more likely to substitute for, rather than complement, the tasks they perform.

*Source: sipea2026061 - References*

---


_Source: https://www.imf.org/-/media/files/publications/selected-issues-papers/2026/english/sipea2026061.pdf_