Bristol, Bath and the wider South West is rapidly positioning itself as world leaders in AI innovation. It’s a very promising time for the region, and thanks to AI Month sponsors, EIP, we were able to conduct a deep dive into the most exciting hives of activity.
AI, or artificial intelligence, refers to using computer systems that can perform tasks requiring human-like intelligence and decision-making. The term is often used adjacently to Machine Learning (ML), a subset of AI. ML means the ability for machines to improve their decision making capabilities by processing data.
“They’ve been influencing our daily interactions for years now”
It’s a common misperception that AI is a shiny new tool, but the term was first coined back in 1955 by researcher John McCarthy. Since then, we’ve come a long way and in recent years we’ve seen AI’s influence infiltrate across the board.
Renowned Bath-based software development agency, Rocketmakers, recently published its AI Guidebook, which I encourage any organisation that intends to implement AI to have a read of.
Within it, author Adam Clarke explains that AI and ML are more than just trending topics: “They’ve been influencing our daily interactions for years now. From the chatbots that guide us on websites to the recommendation systems on our favourite shopping sites, AI is already deeply embedded in our digital lives. These technologies aren’t about creating a futuristic world; they’re about enhancing our current one.”
With the South West driving the national AI narrative, we set out on collating insights from the experts working locally in the field today.
Professor Seth Bullock, Toshiba Chair in Data Science and Simulation summarises, “The area around Bristol has always been a hotspot for companies driven by technology and I think that is continuing in terms of AI, robotics, data science and machine learning.” Consolidated by the announcement of Bristol building the national Isambard AI project, the launch of AI West and the plethora of amazing women founded AI startups, there’s plenty to be excited about here in the South West.
Bristol & Bath as an AI research hub
I couldn’t write about AI in Bristol without highlighting the very exciting news about the UK’s fastest supercomputer being developed here in the city. Professor of Artificial Intelligence, Peter Flach, tells us, “The UK wants to build an AI ecosystem, and Bristol is a key player in that.”
Seth adds, “Over the last year or so, Bristol is probably the leading recipient of significant investment in academic AI research.” The government unveiled that the Isambard-AI would be a £225m investment project; deciding that Bristol should be the birthplace of the UK’s fastest supercomputer is no coincidence.
Alongside this, more funding is coming in from UK Research and Innovation (UKRI) to train the next generation of AI expertise. £117m has been disrupted across the country and the University of Bristol’s UKRI AI Centre for Doctoral Training in Practice-Oriented Artificial Intelligence will oversee 50 PhD students.
“Over the last year or so, Bristol is probably the leading recipient of significant investment in academic AI research”
In addition to this, Bristol will also be leading two of the nine new AI Research Hubs: Collective Intelligence (AI4CI) and INFORMED AI. AI4CI will be developing new machine learning and smart agent technologies fuelled by real-time data streams in order to achieve collective intelligence for individuals and national agencies, whilst the INFORMED-AI hub is developing theoretical foundations and algorithmic approaches for intelligent distributed systems. These systems aim to be effective, resilient and trustworthy in their operations.
Perhaps what best illustrates the big appetite for AI here in the South West is the launch of AI West. Peter tells us this is an open AI community, dedicated to promoting the South West as a centre of excellence in AI, while contributing to the economic growth of the region: “We just had the launch event a few weeks ago, which attracted 100 people – we fully sold out. The energy of the room was really interesting and dynamic, and I’m looking forward to seeing what comes out of the group.”
And the work within local universities has a direct impact on the types of companies we see being founded in the region, and their attitude to embracing the possibilities AI has to offer.
From startups to unicorns
Firstly, we have to mention Bristol’s unicorn, Graphcore. When we asked about key organisations in the sector, both Peter and Chloe Marshall, Head of Marketing at Rocketmakers, referenced the business: “Graphcore in Bristol, focusing on AI processor technology, has been a standout in the region.”
Their total funding now stands at $767M; its largest single round equating to $222M. Graphcore’s Intelligence Processing Unit (IPU) technology enables innovators create the next breakthroughs in machine intelligence to enhance human potential. It’s unique architecture lets researchers undertake entirely new means of work, and has particuarly exciting potential for social impact, from drug discovery and disaster recovery to decarbonisation. Now one of the UK’s leading organisations, their ambition is for their processor to become the worldwide standard for AI computing.
XMOS is another key semiconductor company in the region. A University of Bristol spinout, Their cost-effective products have been previously used to improve the efficiency of new technology. In 2019, XMOS developed a £0.72 voice processor chip that allows Amazon’s Alexa to function by using only two microphones instead of four.
The South West is also dominating the health tech scene, utilising AI to bolster the already life changing outcomes developed by people here. From OKKO Health’s device which monitors macular degeneration and FluoretiQ’s rapid diagnostic tools to Emm’s first-of-its-kind biowearable menstrual cup and Open Bionics integrating AI-powered speech recognition into their bionic arms, we’re not exaggerating when we say there is so much innovation taking place on a daily basis here.
“Much of the really exciting innovation in AI has nothing to do with chatbots or image/video generators”
The sheer scale is impressive. You can read more on organisations harnessing AI in creative ways in our deep tech, health tech and AI listicles across the techSPARK site.
As EIP Managing Associates, Ben Maling & Felix Hall outline: “Much of the really exciting innovation in AI has nothing to do with chatbots or image/video generators. The transformer architecture on which these tools are based has shown itself to be useful in a very broad range of situations, such as in genomics and other areas of biology, the most well-known example being DeepMind’s AlphaFold, which effectively solved the longstanding problem of protein folding in computational biology.
“Robotics is another interesting area: AI is already routinely used in control systems for robots, but is increasingly being used to add reasoning abilities to make them more capable of performing a wide range of tasks.”
If you’re looking to expand your use of AI within your small business, here’s another reminder to check out Rocketmakers’ guidebook. There are a multitude of clever ways you can begin to integrate AI into your business. Creativity is not confined to content. As Rocketmakers outlines, “AI can streamline operations, helping your teams get more done in less time. Whether it’s automating routine tasks, improving workflows, or expediting decision-making, AI has the capability to refine and accelerate numerous business processes. These efficiency boosts aren’t just about speed; they can lead to more informed decisions, fewer errors, and a more agile business overall.”
Getting creative with AI in Bristol
That said, there are also some pretty cool things being developed with generative AI. There’s a good reason the topic has taken the mainstream by storm, and there are some stellar examples of local creative techies utilising it with exciting outputs.
We caught up with Bristol-based marketing agency founder and prodcast producer, David Sloly, to talk about how he’s created the first ever AI sitcom, Made in AI. The show, which was Voted in the Top 5 AI podcasts by Goodpods, follows an artificially intelligent family running riot in the real world, a concept that has been described as “funny and terrifying in equal measure.”
David tells us more: “I quickly discovered that I could make something in two days what would ordinarily take me five days by keeping AI at the core of the process. It’s end-to-end AI involved.”
But what exactly does this mean? David uses GTP-4 for plot ideation, AI Judge to defuse any arguments, Artist IO to create sound effects, Bandlab for producing original music, Eleven Labs to generate realistic speech and MidJourney for all things branding and imagery – to name just a few (check out the full list of tools here).
David recounts utilising Chat GPT to write John McCarthy’s famous speech: “I’ve been working with AI to try and imagine what that speech would have entailed. It makes mistakes. It keeps forgetting that it’s 1955 and there are no mobile phones. But I would have struggled to write it alone, it’s put some lovely bits in there.
“The process is great – I’m having this conversation with something that’s not real, putting it into something that is real, and then getting an email from someone on the other side of the world saying ‘that made me laugh.’ What a strange series of events.”
Whilst producing Made in AI, David realised that some of the tools were remembering the show, his characters and the plot. This fascinated David and inspired him to think about how small businesses might be able to take advantage of this powerful tool. His marketing agency, Harvey David, is in the process of building an AI assistant that creates personalised content for your business.
The rise of generative AI is creating new possibilities for content creation that marketers can monopolise on. Inclued AI, founded by Joyann Boyce, is utilising a similar concept to help teams detect bias in their work. With the increasing integration of AI in the content development process, brands face potential risks due to biases inherent in some AI models.
Inclued AI seeks to solve this problem through its technology, offering peace of mind that your marketing materials – from imagery to copy – are removed from potential stereotypes, harmful tropes and offensive language before publication.
What should we be doing as a community?
During this exciting time, there is lots to be done within our tech community. Throughout our conversations, the importance of education cropped up time and time again. As Chloe underscores, “Promoting AI ethics education and transparent AI use cases in our community is key.”
Joyann adds, “As a community, we need to emphasise that AI is not here to take people’s jobs. It’s to enhance what people are already doing.” She highlights that whilst the umbrella conversations around ethics are of course essential, we should spend more time looking at the responsibilities when it comes to smaller, everyday implementation within businesses.
“Sometimes as a community, we focus a little bit too much on the super high-level things and neglect the small startups integrating AI with no budget. What can they do to implement it responsibly? Providing these resources is key for us to do as a community.”
Within this, Joyann shines a light on the work Laura Gemmell is doing with her startup Taught by Humans. Their ambition is to ensure everyone – both businesses and individuals – are AI & data confident. This means having the skills, knowledge and confidence to be able to use these tools in their work, and if needed their own lives.
The company wants people to avoid scaremongering about AI replacing humans in their jobs, and instead wants to work with businesses to empower their humans to use their tools effectively. You can check out their awesome AI educational tools here.
“We need to work on improving the quality of the discourse around AI and increasing the degree to which members of the public understand the difference between real AI here and now vs. credible AI in the near future vs. science fiction”
Seth corroborates this school of thought: “A key challenge is making sure that the public is informed to the extent that they can make good use of AI and hold AI companies and regulators to account. We need to work on improving the quality of the discourse around AI and increasing the degree to which members of the public understand the difference between real AI here and now vs. credible AI in the near future vs. science fiction. An obstacle is that the conversation can be dominated by individuals and companies with vested interests in hype of various kinds.”
He cites the second way we can continue to make a difference is through, “Targeting AI projects that have the potential for significant positive impact on society, e.g., AI for healthcare, AI for the environment, AI for financial stability, AI for better government. In each case AI has the potential to increase transparency and efficiency and to connect individual people to problems and challenges that can feel overwhelming because of their scale and intractability. AI has the potential to offer a new source of information and assistance for people and policy makers who want to better understand so that they can make better decisions.”
A huge conversation is also being had around AI legislation. We had a fantastic informative chat with Ellen Keenan O’Malley, Senior Associate in the Commercial IP team at EIP about how we’re approaching AI regulation as a country and continent, which you can check out for an in depth analysis.
It’s important we strike the right balance between fostering innovation and protecting the population, as Ben and Felix emphasise, “Many believe AI has the potential to provide profound benefits in terms of productivity, quality of life, and other areas, and therefore argue that we must do all we can to avoid hindering its progress. Others consider that such beliefs are not founded on evidence, but rather marketing, and in fact AI poses a vast range of risks to individuals and groups, and we should prioritise the protection of those individuals and groups. This is a core tension that has to be navigated by legislators.
“There is a broad debate over how and to what the extent AI should be regulated. Some consider the EU’s AI Act, the world’s first top-down regulation on AI, to be overly restrictive and argue that it will unnecessarily stifle innovation while putting European companies at a disadvantage compared with those in the US, China, etc. Others say it is a bold and necessary effort to curtail the many potential harms of AI. In any case, the AI Act will affect all companies providing AI systems in the EU, so may become a de-facto international standard, like GDPR did for data protection (the so-called Brussels effect).”
As we continue to embrace the revolutionary outputs AI has enabled, and act as world leaders in Bristol and Bath, we shouldn’t lose sight of the importance of keeping everyone educated on both the risks as well as the possibilities.
Thanks to all our contributors who aptly highlighted Bristol and Bath as a home to a wealth of AI pioneers. Check out the hub for more related content, give us a follow on LinkedIn to keep your finger on the pulse when it comes to all things techSPARK and sign up for our newsletter for regular updates & content.
