Methodology and evaluation criteria

We evaluated each city across eight core dimensions, assigning a weight to reflect its relative importance in driving AI excellence in urban environments.Individual scores (0–100) for each criterion were multiplied by these weights and summed to yield the final 100-point ranking.

Eight-Criterion Scoring System (100 Points Total)

Research Output and Innovation 20%

Measured by number of peer-reviewed AI publications, patents filed, and presence of leading academic and corporate AI labs.

Investment and Startup Ecosystem 20%

Weighted by total AI venture capital inflows, number of AI-focused startups, and average startup valuation.

Infrastructure Capacity 15%

Assessed via available GPU/TPU compute power (MW of data-center capacity), 5G/edge-network coverage, and cloud services maturity.

Government Policy and Public Funding 15%

Based on dedicated AI strategies, R&D budget allocations, tax incentives, and regulatory support frameworks.

Industry Adoption and Pilot Deployments 10%

Evaluated through scale and scope of live AI projects in healthcare, finance, transport, smart-city operations and public services.

Talent Pool and Education 10%

Indexed by number of AI-trained graduates per year, availability of specialized degree programs, and presence of flagship upskilling initiatives.

Ecosystem Maturity and Partnerships 5%

Gauged by volume of public–private collaborations, international R&D partnerships, and global corporate-research center ties.

Data Availability and Testbeds 5%

Captured by existence of open data portals, national AI testbeds or sandboxes, and volume of accessible real-world datasets.

Total Weight: 20% + 20% + 15% + 15% + 10% + 10% + 5% + 5% = 100%

Data Collection and Validation

Each city's raw performance data were normalized against the top performer in each category, weighted, and aggregated to produce the 100-point composite score used in the ranking.

For each of the above categories, specific indicators will be collected and quantified.For example, for "Research Output and Innovation" we measured peer-reviewed publications and patents filed.For "Investment and Startup Ecosystem" we evaluated the total AI venture capital inflows and number of AI-focused startups.

This process ensures a data-driven, objective ranking.

Scoring Methodology

Once data is collected, we convert raw metrics into scores for each category using the 100-point framework. The general process is as follows:

1
Normalize Metrics
For each indicator, we normalized the values on a comparable scale.Each city's raw performance data were normalized against the top performer in each category.
2
Apply Weights
The normalized scores were then weighted and aggregated.
3
Compute Total Score
The weighted scores were summed to yield the final 100-point ranking.

Determining the Top 10 Cities

The Top 10 AI Cities ranking was created by selecting the 10 leading cities in the world in the field of artificial intelligence.The ranking uses a 100-point scale.

The final ranking is based on the total composite scores, ensuring that cities must perform well across all eight dimensions to rank at the top.

This analysis was performed independently using public AI services.The ratings were obtained through standard user interfaces.The visual design is for informational purposes only to distinguish between sources and does not imply any partnership, official cooperation, or endorsement by the services mentioned.All trademarks belong to their respective owners.