Met Police Commissioner dismisses calls for AI regulation

The Met Police Commissioner has dismissed concerns around the expansion of the use of Artificial Intelligence (AI) in policing – suggesting its continued rollout is the only way to ensure they can keep up with criminals.

The force has been keen to adopt the technology citing limited resources and declining officer numbers, and revealed at the conference "Policing for London in the Age of AI" on 21 July that it would increasingly utilise AI to analyse video evidence in criminal investigations.

But academics and human rights campaigners have suggested that the UK needs both an overarching legislative framework and powerful regulators to ensure that the police procure and use the technology correctly.

However, in an exclusive interview, Met Police Commissioner Mark Rowley said that he was already satisfied with current safeguards, such as GDPR, the Human Rights Act 1998, and the Surveillance Camera Code of Practice.

Rowley told the Local Democracy Reporting Service (LDRS): “There is so much regulation already around data, human rights, surveillance. I don’t think we’re lacking regulation. There’s plenty of that to protect the citizen.

“The bigger risk, and what Londoners have said in recent surveys, is that they worry about [whether] the police can keep up with them on technology – not whether we’re overusing it.”

The Met’s use of AI currently focuses on processing images from around one million CCTV cameras across London. The force said that early use of AI in serious crime investigations has already reduced review time by the equivalent of 454 officer days, freeing up resources and enabling faster progress in investigations.

Rowley was speaking to the LDRS following the summit Policing for London in the Age of AI, which was held at City Hall by and convened by Labour Assembly Member Marina Ahmad.

Senior police officers and advisors joined academics from LSE, Michael Birtwistle from Ada Lovelace Institute and Ruth Ehrlich from human rights organisation Liberty, at the conference.

Earlier, Rowley told attendees that there was “no other option” than to embrace technology, otherwise the Met would risk falling further and further behind sophisticated criminals.

“The world is changing fast, and our ability to catch criminals is degrading because the technology is available to them,” he said.

“We either put fewer people in prison or use technology to help keep pace with the world around us. The big opportunity in policing is exploiting the data we have – both in terms of automating what we do to ensure we have more resources on the front line, and finding the most dangerous people.

“People assume more data means more insight. It was much easier before – now we have all this material, and if you’re really clever, you can stitch together maybe a quarter of the insights needed – but these tasks are getting beyond human.

"If fraudsters are using AI to automate attacks, how can we use AI to counter that – this won’t be a human-scale task.”

Rowley has previously opposed the approach of waiting for legislative safeguards before rolling out technology. Last month, he said: “If every time a new capability emerges, or we find a new use case for data exploitation, we have to wait for new legislation before we use it, we will fail.

“We cannot legislate for every incremental development in technology. The pace is simply too fast, and the legislative process is too slow.”

At the summit held this week, he claimed that the public shared the same vision.

“There’s lots of rhetoric that the public are nervous about this, but the evidence is different to that,” he said.

“The public are actually worried that we are falling behind criminals in terms of using technology. There’s a lot of appetite for this – they expect us to keep up and protect them by making use of technology. That’s not a plea for an unregulated world, but a call for common sense.”

He also called on Assembly Members to “encourage” the Met to go faster on the issue rather than holding them back over ethical concerns.

“The only way we will have a competitive advantage in protecting Londoners is going to be relying on technology,” Rowley added.

Green Party Assembly Member Benali Hamdache, who attended the conference, told the LDRS: “He’s wrong. We heard crystal clear today that there isn’t a framework, there’s a lack of clarity, and we already have examples of the police misusing this technology.

“It’s really important that we do have a legislative framework and that legislators catch up to the reality of what is happening today. Legislation should be an enabler, pointing the police at where use of this technology is appropriate, safe, and constructive.”

The conference heard examples of AI hallucination, with an example given of the research which West Midlands Police’s used in their decision to ban Maccabi Tel Aviv football club fans from a match in Birmingham.

One of the examples that West Midlands provided was hallucinated, ie completely fabricated, in response to their AI prompt.

Marina Ahmad, the City Hall Labour group’s spokesperson on policing, added: “The most important thing for me is that the legislative framework is a reassurance for people, and it will bring, I think, more people along.

“We don’t want things being done to people, we need to bring people along. Otherwise, there will be anger, there will be tension, because people don’t understand what’s happening.”

Sign-up for our free weekly newsletter - Salamander News in your inbox