REGULATING AI

REGULATION

IS IT ENOUGH TO REIN IN THE ROBOTS?

The AI market is a global one and competition is fierce. Encouraging investment is key, but responsible development and deployment of AI also requires innovation friendly regulation. What does that look like?

In Europe, the EU is pushing ahead with new cross-cutting AI specific legislation while the UK Government’s recent white paper confirms that it will continue with its current sector specific approach to regulation.

In this series of briefings, Partner Rob Sumroy and PSL Counsel Natalie Donovan from Slaughter and May’s Emerging Tech team analyse the UK’s policy proposals around AI regulation, and the EU's AI Act which has been agreed on a political level.

FINANCIAL REGULATION

HOW DOES THE CURRENT UK REGULATORY FRAMEWORK APPLY TO AI AND MACHINE LEARNING?

AI technologies have the potential to make financial services and markets more efficient, accessible, and tailored to consumer needs. They are increasingly being used by financial services firms across a range of business areas.

However, AI also raises novel challenges and poses potential new risks to consumers, the safety and soundness of firms, market integrity, and financial stability.  The Bank of England, the Prudential Regulation Authority and the Financial Conduct Authority therefore have a particular interest in the safe adoption of AI in UK financial services, including how policy and regulation can best support it.

The question currently rattling their cage is whether AI can be managed through clarifications of the existing financial regulatory framework, or whether a new approach is needed.  

In this video, Senior Counsel Tim Fosh and Senior PSL Selmin Hakki, consider the challenges associated with the use and regulation of AI in financial services, examining how existing legal requirements and guidance apply, and what is on the horizon.

INTELLECTUAL PROPERTY

IP ISSUES WITH AI DEVELOPMENT AND USE

Countries around the world are jostling to attract AI investment and expertise and position themselves as leading AI nations. An appropriate IP regime will be central to achieving this – with the holy grail being a system that provides appropriate protection for novel AI technologies whilst at the same time striking the right balance between the interests of rights-holders whose content may have been used in training those AI systems and those of the AI developers themselves. Intellectual property regimes are intended to encourage, protect and reward innovation, which in turn propels investment and research in AI, but go too far and the law risks having the opposite effect.

Given that delicate balancing act, are our current IP laws fit for purpose in an increasingly digital, AI-enabled, world? In our latest content piece, we consider some of the key questions surrounding intellectual property law and its interaction with AI, including: does training generative AI using unlicensed third party material infringe UK copyright? Can AI-generated outputs be protected by copyright? Can AI be an inventor for UK patent purposes? And are AI inventions themselves patentable in the UK?

M&A

THE INTERSECTION BETWEEN AI, CORPORATE TRANSACTIONS AND UK NATIONAL SECURITY

Under the National Security and Investment Act 2021 (the “NSIA”), the UK Government has the power to scrutinise and potentially intervene in corporate transactions which raise national security concerns. Recognising the potential national security implications of artificial intelligence technologies, AI is classified as a “high risk” sector under the NSIA regime. This means that corporate mergers and acquisitions in which the target entity undertakes AI activities in the UK will be subject to a mandatory notification requirement.

In this briefing Competition Partner Lisa Wright explains how the NSIA regime operates and how AI is defined for the purposes of this regime. She goes on to discuss the trends emerging from the NSIA’s first year of operation as well as several practical takeaways that parties looking to buy or sell an entity active in the UK’s AI space should bear in mind.

We have also recently published some guidance covering the key issues potential investors should focus on from a legal perspective when approaching AI investments. While these are good practice in any investment scenario, they are particularly crucial when investing in AI - see blog.

GENERATIVE AI

UNDERSTANDING THE OPPORTUNITIES, AND RISKS, FOR YOUR ORGANISATION

It is fair to say that large language models (LLM) and generative AI like ChatGPT and Bard have caught the world’s attention. They have generated headlines in the tabloid and tech press alike, predicting their impact on the way we work, learn and search the internet. User numbers have also increased at an exponential rate. ChatGPT, for example, is one of the fastest growing consumer applications ever, reaching 100 million users a mere two months after launch.

The speed at which their capabilities are developing has, however, raised some concerns. Over 1000 AI experts (including Elon Musk and Steve Wozniak) have called for a pause in the “out-of-control race” to develop ever more powerful AI.

They are also firmly on the radar of regulators and national bodies, who are starting to take steps to help manage potential risks. The Italian data regulator temporarily banned the use of ChatGPT following data privacy and security concerns, and in the US the White House has advised tech firms of their fundamental responsibility to make sure their AI products are safe before they are made public.

In the UK, the National Cyber Security Centre has warned of potential security risks (see our blog), the ICO has produced a set of questions for developers to ask themselves, and the Competition and Market Authority announced it will start examining the impact of AI foundation models (including large language models and generative AI) on consumers, businesses, and the economy (see our blog).

But what does this all mean for your organisation? Do you know how the tech works, the benefits it could bring and the potential risks that will need to be managed?

In our publication Generative AI: Practical Suggestions for Legal Teams, we provide practical tips on how you can use LLMs today and what their development means for your organisation. We also look at what’s on the horizon in this fast changing space.

Our short blog Generative AI: Three golden rules, helps organisations who are, or are planning to use, generative AI to understand the tech itself as well as the opportunities and risks it presents.