Menu
Wed, 15 January 2025

Newsletter sign-up

Subscribe now
The House Live All
Technology
Manufacturing chemical raw materials using captured CO2: pipedream or reality? Partner content
By BASF
Technology
Social affairs
Can ‘Global Britain’ really succeed with electric vehicles? Partner content
By Advanced Propulsion Centre UK (APC)
Transport
Press releases

Tech Secretary Says He Will Take The Flak If AI Expansion Goes Wrong

Peter Kyle has been a Labour MP since 2015 (Alamy)

6 min read

Science, innovation and technology secretary Peter Kyle has said that he will make himself personally accountable if artificial intelligence tools make mistakes across the public sector.

Kyle, the Labour MP for Hove and Portslade, told PoliticsHome that his “instinctive approach” was to be the face of accountability if AI makes errors when rolled out across public services, as it was important for the government to be “open and transparent”.

The government has announced a plan to “turbocharge growth and boost living standards” with the use of AI across Whitehall and the public sector.

The plan will include ramping up the use of AI across the NHS and other public services to allow workers to spend less time doing admin, as well as introducing dedicated AI Growth Zones to speed up planning for AI infrastructure.

In a speech on Monday, Prime Minister Keir Starmer said that "Britain will be one of the great AI superpowers", but admitted there would be "teething problems". With the technology still being developed and improved, there is a strong likelihood that AI tools will make mistakes when used in the public sector.

In an interview with PoliticsHome, Kyle said that during the development of the government's experimental chatbot last year, he insisted on staying close to the process despite officials advising against it.

"It was very important for me to say to officials that I'm not going to stay away from it,” he said.

“I'm going to embrace it, and I am going to be the person that explains to the public what we're doing, why we're doing it, and what could go wrong.

"I said to officials that I will be the person that fronts up anything that comes out of this that could be embarrassing or challenging or could go wrong with this kind of technological development.

“So that gives you an example of my instinctive approach, because the public does need to understand that if we can only train and develop in private, then in this modern digital age, we're not going to get enough interactions to make products and services that are really bulletproof when it comes to these sorts of things."

There are concerns among privacy campaigners that further integration of AI and partnerships between the public and private sectors could lead to data – including information on NHS patients – being used by private companies for commercial purposes. This also raises the question as to whether foreign states seen as national security risks – including China or Russia – could gain access to this data.

Kyle said that patient data could be partly used for commercial reasons, due to the “nature of digital innovation” – but that this would be in a way that “serves the public sector and the public good”.

“We could allow a university that is partnering with a private company or a private investor to have access to health data to create a new medicine, a new treatment, a new innovation that could prevent serious illness or death or disease,” he said.

But he strongly denied that any of this data would be “sold”.

“British data on British citizens will be staying in Britain. We're not going to be wholesale, flogging data off and losing control of it – that is simply not on the table, that will not happen.”

According to Kyle, privacy "will always be the first step", and he wanted to "talk up to the public" during the development of the plan to foster trust in the government.

Keir Starmer annd Peter Kyle (Alamy)

The plan for AI also includes building major data centres across the UK. Government sources acknowledge this infrastructure will likely be very energy-intensive, and Kyle said it "does pose a huge challenge because we're talking heavy infrastructure for us to get it right".

However, the tech secretary said he saw "only opportunity" in their construction.

A new AI Energy Council has been established with Kyle and energy secretary Ed Miliband acting as co-chairs, to “bring the voice of AI into how we make those strategic decisions”. Kyle added that the AI investment zones will be set up in areas where there is already an excess of energy supply.

On the potential threat of AI to jobs, Kyle said the government would ensure they “take communities with us and not leave communities behind”.

“This is something that we really deeply care about. It's very hard to see more than a year or 18 months out which sectors are going to be disrupted, or which parts of the country, but we are doing our absolute best.”

He added that his department was working closely with the Ministry of Housing, Local Government and Communities on this: “We're not going to do what the Tories did, which is stand on the sidelines, and let the market run riot in any way, shape or form that it wants to do and disrupt people's lives in ways that is good for the market, not for communities.”

Kyle takes advantage of AI tools himself, using ChatGPT to learn on the job as tech secretary. He described how he will often have a conversation with ChatGPT “to try and understand the broader context where an innovation came from, the people who developed it, the organisations behind them”. 

“ChatGPT is fantastically good, and where there are things that you really struggle to understand in depth, ChatGPT can be a very good tutor for it.”

Research by the Health Foundation think tank in July last year found that the public and NHS staff broadly support the use of AI in the NHS, with 54 per cent of the public supporting the use of AI for patient care like diagnosing illness and recommending treatment and 61 per cent supporting the use of AI for administrative purposes like sending letters or planning staff rotas.

However, while a third of the UK public thinks AI will improve care quality, a further third think AI will not change care quality. Around 1 in 6 think AI will worsen care quality.

“By the time of the next election, people will be much more aware of AI and how AI is interacting with the services that they use,” Kyle insisted.

“People will be able to access government more swiftly so that they will be more empowered as citizens because they will have information and analysis through which they can make better choices and hold government to account.

“I hope that they will start to notice that efficiency and effectiveness is being driven into parts of the public service that at the moment, are so analog-based that they are literally stuck in the previous century.”

PoliticsHome Newsletters

PoliticsHome provides the most comprehensive coverage of UK politics anywhere on the web, offering high quality original reporting and analysis: Subscribe

Categories

Technology