As Professor of Artificial Intelligence and Robotics and Director of the Oxford Robotics Institute, Nick Hawes is widely regarded as one of the UK’s foremost artificial Intelligence speakers.
His work as one of the country’s leading technology speakers has spanned everything from underwater exploration in Loch Ness to inspection tasks at nuclear sites, showcasing the practical power of robotics in sectors such as healthcare, logistics, and energy.
With a background in computer science and human-robot interaction, Nick is also a recognised figure among Big Data speakers, driving forward the development of AI technologies that integrate autonomy, perception, and intelligent decision-making.
In this exclusive interview with Champions Speakers Agency, Nick discusses the opportunities and risks of AI adoption, the realities of robotic deployment, and what organisations should know about the future of intelligent systems.
Q: As a Professor of AI and Robotics at the University of Oxford, you're at the forefront of emerging technologies. What developments do you believe local and national businesses should be paying close attention to right now?
Nick Hawes: “So there's a lot of really exciting technologies at the moment around both artificial intelligence and robotics. I think for robotics, one of the most exciting things for me is that I think autonomy in robotics is becoming closer to being business as usual. So these are robots that can operate for themselves without direct human intervention, using AI on board to make decisions.
“These are happening in very limited scope, but are typically used for things like logistics, which is quite common now, increasingly for inspection—so having quadruped robots or drones automatically flying around SES, looking for things that have changed, looking for things that might require further inspection from humans.
“So on a robotics perspective, that kind of autonomy is very, very interesting. I think further out there's a huge amount of excitement about humanoids. I think if I was looking to bring robotics into my business right now, I wouldn't be looking at humanoids unless I really wanted to, I think, take some risks. But within the next, you know, five to ten years, there may be some use cases for humanoids.
“I think if you look beyond that to the broader AI scope, there's just huge excitement around foundation models—so large language models, large vision and language models, things that effectively compress all of the knowledge of the internet or of specialised data sets into something that you can query very quickly.
“People in robotics are using that to understand the scenes around robots so they can interact with the world or humans better, or just to give robots more general capabilities to act in otherwise unstructured environments.”
Q: With AI becoming increasingly embedded in workplace systems, what are the key advantages and risks that organisations should be considering?
Nick Hawes: “Perhaps the biggest con is that we don't—you know, probably we don't know how to use AI very well. We don't really understand some of the legal aspects, things like copyrights. So there is quite a risk in introducing this into your workflows.
“Honestly, I think one of the biggest scary things to me is the energy requirements right now. So anyone using AI is really contributing to kind of the climate crisis. I mean, we all do it, we all use a whole bunch of electronics, but the kind of training and inference energy cost of AI is something that people tend to overlook.
“So I think when you're looking at your kind of carbon footprint as an industry, I wonder—I'm curious to know how AI is incorporated into that. But I think people are getting good at dealing with what are sort of some of the more widely known downsides of AI, about the kind of hallucinations, the unpredictability.
“So there's lots of people that look at kind of how do you focus the use of AI, particularly kind of language models in particular ways and constrain their output to, let's say, reasonably predictable areas.
“And I think that's where those kind of uses—when you think about chatbots, when you think about data retrieval, when you think about prototyping visual designs, code, documents—that's the real win. So previously, a lot of these tasks were completely—not impossible to automate, but very difficult to automate—and the kind of AI we're seeing now allows us to automate a broader range of tasks.
“For example, querying large unstructured documents, interacting with customers on kind of, let's say, very specific topics. So we can do a range of tasks and we can also do them in a much more general form.
“So if you think back to the automation that we might have had five or ten years ago, in terms of chatbots, in terms of scripting of apps, these things would have been very—or, and a lot still are—very rigid, very structured. You can only interact with the system in a particular way, you can only control its output in a very particular way because those are the ways that humans have decided that that should work.
“But the advent of these large AI models allows a greater range of flexibility and generality within a particular task, and means that the input can be a lot less structured and the output can be a lot more controlled.
“So there's, I think, real advantage in the kind of approaches we see now to be able to tackle problems that just couldn't be tackled before. But maybe the other downside is we shouldn't get too carried away. These are all still kind of single-shot processes typically.
“It might be a single dialogue which has multiple steps or a single image generation, but there's not a lot of systems that can actually more autonomously complete a series of separate tasks to achieve a goal.
“So you think like booking a holiday or arranging a delivery—these things are more than just a single step or a single dialogue. You have to have multiple independent parts that are coordinated. I think that's one of the areas where current AI systems are lacking—that kind of planning and coordination capability across multiple domains.”
Q: You've deployed autonomous robots in a wide range of environments—from care homes to nuclear facilities. Can you share some of the most notable projects you've led?
Nick Hawes: “Over the years, I've deployed autonomous robots in a range of different places—quite kind of an extreme range, I think. Some of my earliest work looked at deploying autonomous mobile robots, so wheeled robots, in indoor settings. There, we put robots into offices doing kind of security and patrol tasks, and also into care homes or hospitals, where the robots were supporting nursing staff.
“These robots operated autonomously for months at a time with no expert human intervention, so they were kind of truly autonomous and capable of doing a small range of different tasks. Since then, I've deployed robots all over the place. So we had an underwater robot operating autonomously in Loch Ness—this is with colleagues here at Oxford and in the National Oceanography Centre. This was an underwater robot collecting data from a network of sensors.
“We've had robots operating in radioactive environments—so operating around the outside of the JET fusion reactor in Culham down the road, as well as doing inspection tasks in Sellafield, autonomously inspecting the Calder Hall power plant under decommissioning. And we've also applied robots into forests, into grasslands—kind of across the board really. So, you know, everything from care homes to nuclear reactors—I've had robots operate autonomously in those.”
This exclusive interview with Nick Hawes was conducted by Mark Matthews.
For More Information: Champions Digital Disruption Speakers