In the end, trust will win out

The challenges facing AI in Europe – and what PROWLER.io is doing to stay ahead.

back to our blogs

In the end, trust will win out

The challenges facing AI in Europe – and what PROWLER.io is doing to stay ahead.

Thousands of visitors this week braved a deluge of rain – torrential even by the standards of London – to attend CogX 2019, Europe’s leading AI conference. It was a fascinating event, and I’m delighted to report that on Monday night the organisers of CogX gave us a special award for our work: we were winners in the 'Best Innovation in Simulation' category. These awards are intended to celebrate what the organisers describe as "the best-of-the-best in AI and emerging technologies".

We’re honoured, but I can’t say we’re surprised. As co-founder and CEO of PROWLER.io, I am convinced that here in Europe, not just in London, we can lead AI development globally. There is a vast potential for investment here, and we and many others are doing extraordinary things. We just need to be more confident and forthright about it.

At the conference itself, among the keynote speakers was London’s Mayor, Sadiq Khan who arrived at the event to open the conference and launch London Tech Week which coincides with CogX. He was unequivocal in his support for intelligent, considered AI, blaming Brexit for side-lining preparations for the long-term future of AI. I’m inclined to agree with him.

He said that the political landscape surrounding Brexit was "draining central government resources and attention away from one of the most important debates of the day." "There's no doubt that as a country we should be much further ahead than we are now," he said. "Not only in terms of making sure we can make the most of what AI has to offer, but in terms of educating the public, encouraging public debate and preparing for the potential impact on our society."

He’s absolutely right – we in Europe need to own AI. Especially, trusted and ethical AI. If you read the news and read the popular press, you might be forgiven for believing that here in Europe we are stuck in some sort of virtual world – parked, waiting impotently for something to happen. Here we are, so the myth goes, languishing in a no-man’s land between China and the US. True, the United States and China together have a population of 1.7 billion people, and that leaves about six billion of us in the middle. But we are not sitting around twiddling our thumbs.

PROWLER.io is a small company in Cambridge but in terms of research we are definitely punching well above our weight. In 2018 we came 6th in terms of the numbers of papers published at NeurIPS. We were ahead of everyone other than Google, Microsoft, IBM, Facebook and Intel. And if you look at it another way – we are getting a massive bang for our buck. We spend only a tiny fraction of the amount on research that is spent by the big corporations – and we have a particularly impressive hit-rate in terms of NeurIPS. So far in NeurIPS and elsewhere we have published a total of 44 peer-reviewed papers - and there are more in the offing.

The quality and scale of that research has an obvious trickle-down effect – it will help secure the commercial future, through ethical and trusted AI, of our clients. We have to look at these results in context. Building efficient solutions has huge implications for the future of AI. Size is not everything; more important is a co-ordinated approach.

Look at what happened in Europe in the 1980s with the advent of GSM: it was a joint, considered initiative between European technology companies and European governments. It made sense and it was co-ordinated.

Within AI at the moment, there is a huge amount of fragmentation. That needs to stop. We need to find a co-ordinated, ethical way of making the most of the ideas we have and the talent at our disposal. We need to work together with our commercial partners to create solutions that will benefit everyone.

We are actually entering the third phase of AI – where we are looking at how the human mind can be teamed with artificial intelligence to produce the best results. Companies of all sizes are looking for a technological breakthrough that is powered by human intelligence – and empowered by AI. But it all needs to be based on trusted foundations. With the help of our Nordic partner Mandatum Life we are looking at ethical AI for asset management; in the logistics field, we are coming up with solutions that will save millions of dollars in the supply chain and massively reduce the carbon footprint of our clients’ vehicles. And in education, we are also rolling out new ethical AI programmes which will have a huge benefit for educators – and the children they are entrusted to teach.

So much of the conversation, so many of the keynote speeches at CogX have revolved around the issue of ethical AI. My colleague Sarah Jarvis, in her panel discussion with Eric Daimler, an expert in robotics and adviser to President Barack Obama’s administration, touched on the far-reaching implications of Trusted AI and why it is so important. In the end, it is all about Trust.

Ethical AI needs to be built into our systems as part of business performance. Trust is a fundamental requisite - without it, our AI strategies will simply fail, and so will our businesses.

When l think about our future, and what AI can contribute to that future, I am often reminded of Rachel Botsman, who wrote the definitive book on ethical technology, Who Can You Trust? How Technology Brought Us Together – and Why It Could Drive Us Apart. She was unequivocal about the fact that Trust is essential in AI and technology. It’s not a nicety, it’s not an optional extra. Our very lives, our jobs, our families depend upon it. We need to keep that at the forefront of our thinking.

Join us to make AI that will change the world

join our team