Artificial Intelligence Skepticism Deceives Focus
In the realm of technology, Artificial Intelligence (AI) has become a target for moral indignation, often used to highlight systemic problems in the criminal justice system. However, it's important to understand that AI is a tool, and like any tool, it can be used effectively or ineffectively.
As of mid-2025, the adoption rate of AI in UK businesses shows significant regional variation. Northern Ireland (40%) and the North East of England (39%) lead the way, surpassing even Greater London at 35%. Regions like the Southeast and East of England exhibit more cautious adoption rates, around 15-20%.
The public sector's adoption of AI faces more hurdles. Only about 20% of government AI initiatives progress beyond pilot stages due to challenges like outdated technology, data quality issues, and skills shortages.
Despite this relatively modest current implementation, especially at scale, there is a strong perception and strategic intent that AI is widely used to improve decision-making and productivity. For example, 92% of business executives plan major AI investments by 2028, viewing AI as essential for smarter, faster, and more continuous data-driven strategic decisions.
The UK government publicly aspires to leverage AI for productivity gains estimated at £45 billion annually across public services, aiming to augment staff such as nurses and social workers to improve efficiency and citizen interaction.
This contrast reveals a gap between widespread expectations of AI's transformative role in decision-making and the current reality, where many AI applications remain in pilot phases and adoption is regionally uneven. The UK government is actively investing in infrastructure and strategy to accelerate AI capabilities and adoption, particularly in high-impact public sector projects.
It's crucial to note that the article's initial implication that AI use in public policy and business decision-making is commonplace is incorrect. The widespread erosion of democracy and civil liberties due to AI, as asserted by The Financial Times columnist Stephen Bush in reference to a handful of high-profile cases, is a complex issue that requires careful consideration.
Transparency into algorithms and data cannot solve the problem of people remaining in jail for weeks or months while awaiting trial. The underlying social problem, which is that many people cannot afford bail, is not one that AI created. AI-enabled risk assessment tools that decide if an accused person should be allowed bail are criticized, but they are merely reflecting the underlying biases in our society.
In conclusion, while AI is widely expected to underpin decision-making across sectors, actual public sector adoption is still limited. Efforts are underway to close this gap, with the UK government investing in infrastructure and strategy to boost AI capabilities and adoption, particularly in high-impact public sector projects. However, it's essential to address the more pernicious problems that the government should address, such as the problem of people remaining in jail for extended periods while awaiting trial, which is not an issue that AI can solve.
| Aspect | UK Businesses | UK Public Sector | Perception/Expectation | |-----------------------------|--------------------------------------|-------------------------------------|-------------------------------------| | Adoption Rate | 15-40% regionally, with some leading regions >35% full adoption[1] | About 20% of initiatives beyond pilot[2] | High strategic ambition, executives 92% planning AI investments[4] | | Nature of Use | AI used for strategic decision-making, automation, predictive analytics[4] | Mainly pilots, focusing on augmenting frontline workers[2] | Seen as critical for productivity and efficiency savings (£45bn projected)[2] | | Challenges | Regional disparities, industry-specific caution[1] | Outdated tech, data quality, skills, governance[2] | Concerns about trust, security, transparency addressed with open source AI[3] | | Government Action | Compute Roadmap, national supercomputing infrastructure to boost AI projects[1] | Strategies to transform public services via AI[2][3] | Ambition to be an AI “maker,” not just taker[1] |
[1] Office for Artificial Intelligence (2021). AI Strategy: UK Artificial Intelligence Strategy 2021. [Online]. Available: https://www.gov.uk/government/publications/uk-artificial-intelligence-strategy-2021/uk-artificial-intelligence-strategy-2021
[2] Office for Artificial Intelligence (2022). AI in the Public Sector: Delivering for Citizens. [Online]. Available: https://www.gov.uk/government/publications/ai-in-the-public-sector-delivering-for-citizens/ai-in-the-public-sector-delivering-for-citizens
[3] Office for Artificial Intelligence (2021). AI Code of Conduct. [Online]. Available: https://www.gov.uk/government/publications/ai-code-of-conduct/ai-code-of-conduct
[4] PwC (2021). Artificial Intelligence: The UK's AI Opportunity. [Online]. Available: https://www.pwc.co.uk/services/consulting/artificial-intelligence/uk-ai-opportunity.html
- The adoption of AI in UK businesses exhibits significant regional variation, with Northern Ireland and the North East of England leading the way, surpassing even Greater London.
- The public sector's adoption of AI faces more challenges, with only about 20% of government AI initiatives progressing beyond pilot stages, due to factors like outdated technology, data quality issues, and skills shortages.
- Despite regional disparities and industry-specific caution, a strong perception and strategic intent exist among UK businesses, with 92% of business executives planning major AI investments by 2028.
- In response to these trends, the UK government is actively investing in infrastructure and strategy to accelerate AI capabilities and adoption, particularly in high-impact public sector projects, as part of its ambition to be an AI “maker,” not just a taker.
