Not just Chat, the AI applications for GPs and Transport

I think we’ve all got the message by now that Artificial Intelligence is going to revolutionise the world, even if some of us knew this before Chat GPT.

Finding the applications outside of doing the kids homework and writing labourious emails can be a little bit of a challenge. But at the heart of the current artificial intelligence boom is the simple fact that this is all built around data.

As Emeilia Molimpakis, co-founder and CEO of Thymia pointed out at the recent London Tech Week, this simple data processing can lead to huge efficiencies in the healthcare field, as algorithms and software can pick up on longitudinal trends better than doctors.

“Our AI models analyse biomarkers and people’s speech patterns, their facial expressions, and their behaviour to analyse symptoms and patterns with depression, anxiety, ADHD, and many more conditions.

“We gather data from people through gamified activities or mental video games to actively monitor patients, put those with depression and anxiety at ease, and to help clinicians monitor them better in between therapy sessions before the assessment.”

Utilising Data

Molimpakis’ experience in healthcare is indicative of the best uses for data that is being generated.

The usage of this data is also something that the transport sector is looking into, as it continues to move towards autonomous vehicles. As Alex Kendal co-founder and CEO at Wayve, also on the panel at London Tech Week explained, the use of AI is firmly placed to help make safer journeys.

“In transportation, autonomous vehicles in particular are a chance for us to transform the safety, the accessibility, and the way we move people and goods around our cities. The UK loses 2% of its GDP each year due to road accidents and there are 4,000 deaths around the world due to road accidents, so there’s an extraordinary opportunity to improve lives and benefit the economy.

“What I’m excited about is, thinking about the last 10 years of the development of these technologies, the focus has been on amassing the largest amount of computing data around a specific problem vertical.

“The transformation this year with our foundation models is the ability to leverage more general purpose machines, ones that can transfer knowledge from one use case to the other and be guided by prompting. I think this is the really exciting thing because it means you can now bring together multimodal data sources for different applications.”

Public view

Despite the way that, frankly, AI can help us stay alive, the public perception of technology can be a barrier for these sectors to overcome.

However, as Kendall mentioned, the goal is to assist rather than replace, with Viscount Camrose, Parliamentary Under Secretary of State for the Minister for AI and Intellectual Property at the Department for Science, Innovation, and Technology, adding that “AI is a topic of enormous excitement, one particular Founder CEO said to me ‘the money does not exist with which I could have bought that kind of marketing to generate the kind of excitement in my product’ when he was talking about ChatGPT.

“But generating trust is, in my opinion, the number one priority from a government perspective. I think there’s one point I’d like to stress here which is how do we make people trust AI?

“I suppose it’s almost difficult to imagine two industries where there would be greater levels of anxiety, certainly we’d hesitate to get into a car that was gonna drive itself, and you would hesitate to see a robot doctor. Regulation is a very big part of building that trust, but it’s something we have to do delicately, accurately and quickly, because the technology is running away very quickly.

Regulation

As Camrose says, the answer to these concerns seem to naturally be to put some rules in place in order to make sure that businesses, or the AI itself isn’t getting out of control.

As a result, Camrose said that the government is taking a sector-specific approach, in which different industries can expect different rules, adding: “To my mind there are three fundamental design principles around trying to regulate AI.

“The first is don’t try and go after specific instances of technology because technologies can change much more quickly. Second, don’t attempt a cross sector regulation. And third, there is no way to predict in advance what the future is going to look like and therefore the regulatory model needs to be adaptable to change.”

Both Kendall and Molimpakis agree that this approach is the right one to take, with Kendall saying: “I think the approach of having sector-specific regulation is a really important one. If you draw an analogy to how we regulate accountants, doctors, or teachers, it’s all very different, the risk profile is different, the opportunity and the challenges are different.

“I don’t think we’re at the point where we have general purpose AGIs (Artificial General intelligence) that are across the board, so making sure that we individually target healthcare, for example, to make sure that we have regulation that accelerates the opportunity in that space in a way that’s safe and trusted.

“I think [the UK] has an amazing opportunity to be best in the world” said Molimpakis, “I think the government’s approach looks like it’s going in a sector specific direction which is really important.

“But also the UK has one of the world’s best institutes, the Alan Turing Institute, who always had this idea about ethical approaches to AI. We have the foundations to do this properly, and to do it in a way that is accelerating instead of hindering smaller and bigger companies.”