AI adoption has been identified as a key growth strategy for the UK. The UK government has accepted the ideas and plans outlined by Matt Clifford (tech entrepreneur and Chair of the Advanced Research and Invention Agency) in his AI Opportunities Action Plan.

The government is committed to investment in AI and here Matthew Lingard and Paul Armstrong comment on the key points coming out of this commitment.

• Infrastructure: the plan calls for increased investment in AI infrastructure and computing power. It identifies that investing in data centres is key to providing the computing power needed to develop frontier AI technologies.
• Access to data: AI requires data on which to train and learn. There’s a recognition in the Action Plan that many valuable data sources are locked, particularly those under the control of the public sector, and that there needs to be a way to unlock this valuable resource in a safe and ethical way to develop cutting edge AI systems. It will be interesting to see how this is developed, as data is seen as a valuable asset and won’t be given away cheaply.
• Skills and education: there’s a need to develop AI skills in the workforce and an acceptance that educational bodies will need to provide more training and courses.
• UK at the forefront: The UK government doesn’t want the UK to just be a buyer and user of AI, it wants it to be a “maker” as well. There’s an eagerness for us to be at the heart of development of frontier AI and clearly there will need to be significant prioritisation and resource to achieve this.

The Action Plan stresses the need for ethical guidelines and robust regulatory frameworks so that AI technologies are developed and used responsibly. Underpinning all of the ambition of the Action Plan is a requirement for a clear understanding of how AI models can be trained and used lawfully.

We’ve seen a sophisticated piece of legislation enacted in Europe which establishes a regime for the development and use of AI. While we’re still waiting for important guidance from the EU on the practical application of some aspects of the EU AI Act, it does seek to create some levels of certainty around AI development and use. Contrast that with the UK which is taking a wait and see approach to regulation, on the basis that until we understand the full potential of these technologies it’s hard to regulate them.

There’s merit in that approach of course. But too much uncertainty around how IP and data protection laws apply to the training and use of AI models may stymie growth in some of the frontier applications of AI. If the government wants to enable the ambition outlined in the Action Plan then quickly settling some of the uncertainty in the areas of IP and data protection will be vital. Matt Clifford does acknowledge this in the Plan.

The implementation of the UK government’s AI Action Plan will undoubtedly shape the future of AI development and use. But the uncertainty around IP and data protection laws must be addressed quickly to prevent slowing down innovation. If your business is looking to leverage AI’s potential, then staying informed about the evolving legal and regulatory landscape will be key. Read our guide on demystifying AI here or join our webinar as we discuss the “EU AI Act: Regulatory requirements and risk, is your business ready?”. Please get in touch with Matthew or Paul if you have any queries or need further support.