What is your broad-brush definition of artificial intelligence (AI)?

It’s best to think of AI as a technology designed to replicate human behaviour. It’s new and exciting, but it is just technology that allows us to interrogate large amounts of data and come up with ways to use that data to solve problems and provide solutions.

If you’ve been around for as long as I have and started your working life using big data centres and green screens, followed by laptops, desktops and servers and now cloud computing, it’s just another technology.

Looking at AI from a legal perspective, the European Union has issued some AI rules and guidance. Can you outline what they cover and where the UK stands?

The EU AI Act came into force in August 2024 and it’s effectively a two-year staged piece of all-encompassing regulation. It’s the first regulation of its kind in the world and attempts to effectively ensure that consumers are kept safe when manufacturers and deployers use AI systems. The regulation is based around product safety legislation, so there’s a lot of GDPR requirements in its defining structure.

The UK is looking at the regulations and trying to understand how they apply to its jurisdiction and whether to follow the EU AI act or replicate it in some way.

However, the EU has an extraterritorial impact and so manufacturers and deployers of AI systems who interact with consumers within the EU, but who are based in the UK, will be affected by the regulations.

If you are a UK retailer and only dealing with UK clients, not utilising AI at all with consumers within the EU, then you probably fall outside that regulation.

From a UK perspective, the current government and the previous one, both made it clear that AI is going to be an important linchpin of the economy moving forward with the UK positioned as being a leader in the way AI systems are safely adopted. So, the government will have to regulate in some way, shape or form.

The government is looking at how the different industry sectors are regulated to see how AI can be regulated by sector. That’s different to the EU AI Act, which is a piece of horizontal legislation which covers the technology.

What’s the best approach for car retailers to adopt with regards to AI?

The starting point is to understand that this is just technology. It’s very clever, but it’s still only technology and technology is an enabler that allows businesses to do things in different ways.

You need to identify the problem you’re trying to solve and how AI is going to help customers. You also need to understand how you’re going to go about contracting for the use of that technology, because you’re either going to buy in from a third party or develop it yourself, using your own employees or third party contractors. So there’s a contractual matrix around that.

Retailers should consider their use of data protection principles and how they have dealt with the acquisition, concentration and organisation of personal data from their customers, and use that as a foundation.

If a car retailer is not using AI today and they’re considering engaging with a supplier, how do they ensure the data is not going to be leaked elsewhere?

They need to clearly identify the problem or opportunity that AI can address and make sure they read the terms and conditions to ensure that the new technology is covered by what is said within it.

From an OpenAI ChatGPT perspective, there are tick boxes you can select in the application that say that OpenAI will not use the data you input. Those types of clauses make it very clear that you will not agree to the data being outside of the specified purposes. It’s a contract.

As AI tools become more widely used, car retailers need to be aware that while no specific regulations around the technology currently exists in the UK, they should be aware of how regulations are guiding usage in the EU and nsure that GDPR continues to protect the use of all customer data.

Watch out for Part two of our latest AI Auto Talk special – coming soon!