The UK’s AI Standards Hub has strengthened international engagement and improved access to technical standards, although an independent evaluation has identified difficulties converting that participation into sustained collaboration and measurable business adoption.
The assessment found that 77% of surveyed participants believed the Hub helped give the UK artificial intelligence community an internationally recognised voice, while 91% agreed that it improved understanding of how technical standards influence regulation.
Published by the Department for Business, Innovation, Science and Trade, the evaluation examines activities between April 2023 and March 2025.
The Hub was established in 2022 to help organisations understand, navigate, and participate in international AI standards development.
Its work has involved the National Physical Laboratory, British Standards Institution, and Alan Turing Institute, with government support. The UK Accreditation Service is also a partner in the current arrangements.
The programme provides technical information, training, research, networking opportunities, and access to the Standards Observatory, which brings together information about relevant standards and policy developments.
Its second phase concentrated particularly on expanding international participation and strengthening connections between organisations involved in AI governance.
The evaluation found progress in both areas. More than 60% of interviewed stakeholders reported improvements in awareness and visibility, while over 80% highlighted the Hub’s contribution to participation beyond established technical communities.
Some 81% of surveyed participants believed it represented the UK internationally.
These findings reflect respondents’ assessments of the programme, rather than independently measured changes in international standards decisions or the number of businesses adopting particular requirements.
Benefits remain concentrated in information access —
The evaluation estimated that the Hub generated between £1.04m and £2.7m in productivity benefits over the two-year assessment period, compared with operating costs of approximately £2m.
Those benefits mainly represent estimated time savings associated with finding technical information and coordinating engagement. They do not measure productivity improvements achieved through deploying AI systems within businesses.
The evaluators found the strongest evidence of value in the Standards Observatory and the programme’s activities bringing together organisations working on technical standards.
The Observatory reduces the effort needed to identify relevant publications, while workshops and international events provide opportunities to discuss how standards are developed and applied.
These functions can be particularly useful to organisations without dedicated standards specialists, including smaller businesses attempting to understand changing technical and regulatory expectations.
AI standards can provide frameworks for risk management, organisational governance, testing, and assurance. However, technical standards are not automatically legal obligations, and participation in their development does not demonstrate compliance with a particular regulatory regime.
International collaboration has become increasingly important as countries develop different approaches to AI regulation and seek greater consistency in how systems are assessed.
The Hub’s activities have supported discussions between technical specialists, regulators, researchers, and businesses, helping reduce fragmentation across communities that might otherwise operate separately.
Nevertheless, the evaluation found limited evidence that initial engagement consistently developed into sustained cooperation or direct adoption of standards.
Longer-term outcomes, including regulatory alignment and changes in business practices, were still emerging during the assessment period.
The evaluators recommended stronger follow-up arrangements, more structured working groups, improved monitoring of repeat participation, and clearer measures of progress towards collaboration.
They also recommended maintaining the Standards Observatory, updating training as regulatory requirements change, and establishing better evidence of how participation influences subsequent decisions.
The assessment used monitoring data, surveys, stakeholder interviews, and economic modelling. Its estimated financial benefits depend partly on assumptions about time savings, while its wider international impact findings draw on participants’ reported experiences.
The government has committed to continued cooperation with partners on AI and digital standards policy. The evaluation provides evidence of improved access and international engagement while identifying the further work needed to establish whether those activities produce lasting changes in business adoption and technical governance.




