Thoroughly knowledgeable,
very pragmatic and
quick-thinking

Chambers Guide

Insights

Keep up to date with our latest insight pieces, news and industry developments. See below for the latest posts or use the categories to hone your search for stories of interest.

Rather listen? The WABChats Podcast provides engaging and informative conversations with contacts, clients, advisors and friends of White & Black Limited. Listen Now.

Artificial Intelligence – Regulatory Update – Part 3

In part three of our Mini-Series, our team take a closer look at the effects of AI Regulation in the EU and UK’s regulatory approaches.

In our AI-Mini Series, our commercial team take a look at recent regulatory developments in the AI space, what the effects of regulation will look like and expected timelines for enforcement.

Part two of our WABComment series looked at the UK’s regulatory approach to AI.

What is Artificial Intelligence?

Generally speaking, AI is a system this is trained upon specific data to analyse and detect patterns, which is then used to detect or replicate those patterns in new data, such as interpreting speech or spotting specific items in texts or images. There is not yet one single legal definition of AI, although as we will see below EU regulators are seeking to set a definition for their purposes.

What are the effects of AI Regulation?

EU Regulation
Each of the risk classes set out above impose different levels of compliance obligations.

High-risk AI systems will be subject to strict obligations including:

  • Adequate risk assessment and mitigation systems;
  • Use of high-quality datasets feeding the system, to minimise risks and discriminatory outcomes;
  • Logging of activity to ensure traceability of results;
  • Detailed documentation requirements, providing all information on the system and its purpose necessary for authorities to assess its compliance;
  • The provision of clear and adequate information to users;
  • Appropriate human oversight measures to minimise risk; and
  • High levels of robustness, security, and accuracy

Limited risk AI systems will be subject to specific transparency obligations; and

Minimal risk will not be caught by the AI-specific regulation.

Each of the levels of risk under the EU’s AI Regulation carry different levels of potential penalties. Breaches of:

  • unacceptable risk obligations can incur a fine of 6% of global revenue or €30m; and
  • high risk or low risk obligations can incur a fine of 4% of global revenue or €20m.

 

UK Regulation

The White Paper makes it clear the responsibility for the day-to-day operation of AI regulation will sit with existing regulators and on this basis it is likely that specific regulators will introduce new AI-specific ongoing management, record-keeping and audit requirements. It is worth noting the UK government has not specified any potential consequences or penalties for breaches of the AI regulations. It would seem for the moment any penalties would be the same as those that can currently be imposed by the relevant regulatory body governing a particular industry.


The ICO has recently updated its guidance on data protection in relation the use of AI, while also stating the guidance would need further updates to keep up with the pace of technological developments in AI.

 

Other Laws

No other legislative changes are currently proposed, though practical use of AI may come into conflict with existing law such as in relation to health and safety, and data protection.


Our AI-Mini Series was produced by Richard WilkinElla Paskett and Emily Read. Please reach out to one of the team for more information on AI, or find out more about how we can help you.

Disclaimer: This article is produced for and on behalf of White & Black Limited, which is a limited liability company registered in England and Wales with registered number 06436665. It is authorised and regulated by the Solicitors Regulation Authority. The contents of this article should be viewed as opinion and general guidance, and should not be treated as legal advice.

This site uses cookies to improve your user experience. By using this site you agree to these cookies being set. To find out more see our cookies policy