Artificial intelligence

Managing risk in a
rapidly developing
landscape

Machine learning and generative “AI” tools offer the potential to create enormous efficiencies in how we find, analyse and present information. The prospect of using a machine “trained” on a vast quantity of data to produce actionable, plain English responses to inputs and questions opens up a world of possibilities.

While these tools have been around in various shapes and forms for a while, it’s only in the last couple of years that they have really attained critical mass as a mainstream commercial proposition with practical applications.

The legal and contractual landscape around these tools is still relatively immature, although it is evolving rapidly. There are numerous live legal claims and regulatory investigations against providers of these tools for alleged IP infringement or data protection breaches. The UK government is actively looking at the issue, and our neighbours in the EU are close to enacting the first major law specifically regulating AI.

It follows that the prospect of implementing these tools in your business may be exciting, but it can also be daunting. What risks do these tools present in your use case? What can you do to protect or mitigate against them? What happens to your data once fed into the system? What should the contract for providing the system look like?

We can help. We’ve been advising businesses on procuring and implementing these tools, and the governance around their use, for pretty much as long as they’ve been on the market, and we have decades of collective experience in advising on procurement and implementation of new technologies more broadly.

Related services