What national AI plans get wrong and how to fix them | Brookings

Many nations are developing artificial intelligence (AI) plans, but a common mistake is trying to build a complete “AI stack” – compute, chips, models, and regulations – without considering their existing economic strengths. The article argues that a more effective strategy involves strengthening a country’s existing industries, workforce, and institutions, and then using AI to enhance and diversify into related, higher-value activities. AI deployment should be tailored to a nation’s unique capabilities and should focus on building “cognitive infrastructure” – the connection between data, expertise, and systems – rather than solely on compute capacity.

Analysis of global AI investment reveals a geographically and sectorally distributed landscape, with different countries specializing in different AI applications. The U.S. has broad investment, while Europe focuses on enterprise software and medical AI, India on fintech and education, and East Asia on manufacturing. This specialization isn’t random; it reflects each region’s economic DNA and existing assets. Countries should identify their comparative AI advantage and focus on areas where AI can provide the greatest enhancement.

The article proposes a roadmap for countries to leverage AI by first amplifying existing strengths and then strategically diversifying into adjacent sectors. Examples include Norway using AI to improve offshore energy, Germany enhancing its automotive industry, and Bangladesh enabling smart textiles. Ultimately, the most successful national AI strategies will focus on deploying AI within existing economic frameworks, building cognitive infrastructure, and fostering specialization rather than pursuing generic AI leadership.

(Source：Brookings)