of its inputs and outputs. As previously mentioned, there are ethical concerns linked to the way AI uses learned data that can perpetuate stereotypes. Leaders should consider what they need to do to mitigate this, from both a customer-facing and internal perspective. They must also think about whether there will be legal or regulatory implications linked to the way they use the technology. For example, when it comes to generative AI – which creates a wide variety of data that includes images, videos and audio – there are still questions around who ‘owns’ the content produced. Businesses need to make sure they protect themselves and the work they are doing – and, perhaps more importantly, ensure they are protecting their customers. At AMBS, we are involved in a vast range of research and business engagement projects linked to AI, including the Centre for Digital Trust and Security; the Turing Innovation Catalyst; and the Centre for Doctoral Training (CDT) in Decision Making for Complex Systems. All of these initiatives are designed to create the interdisciplinary teams required to address current AI challenges and support the successful translation of AI technology. It is important for us to bring that holistic approach into our teaching, drawing on a range of expertise and different disciplines so that we can equip students with the skills not only to use AI, but to analyse and, where needed, question its value when it comes to ethics, legality and other practical applications. Considering practical applications An interdisciplinary approach has been vital to the setup of the new CDT in Decision Making for Complex Systems, which aims to provide students with a rounded knowledge of AI and, importantly, an understanding of how it can be effectively applied in complex science and business scenarios. The CDT is funded by UK Research and Innovation and jointly governed and delivered by researchers from a range of disciplines at the universities of Manchester and Cambridge, including two academics from AMBS. Students within the training centre will spend the first year of their degree in Manchester, receiving teaching from both institutions, before transitioning into three years of full-time research in either Manchester or Cambridge. There will be annual conferences with the entire cohort, so that students can share what they have learned and get a chance to see the other work
going on at the CDT. Creating these knowledge-sharing platforms is key to giving students a rounded understanding of AI. Research projects within the doctoral training centre fall into a wide range of areas, with a specific focus on advancement in AI methodology and theory; applications of AI in biology and health; applications of AI in physics and astrophysics; applications of AI in engineering; and the translation of AI into business contexts. While directorship of the centre sits in computer science, the management board and supervisory pool brings together researchers from a wide spectrum of disciplines, from across the University of Manchester’s three faculties, to ensure we can develop innovative solutions that are appropriate to the complexity and uncertainty of real-world systems. The involvement of AMBS additionally underpins some of our close engagement with industry partners, linking directly into projects, as well as mentoring and training opportunities. Students within the CDT may come from any of the underpinning disciplines, meaning their PhD will foster knowledge and skills in developing AI systems and train them in applying this knowledge to the benefit of their chosen domain. In terms of supervision, the majority of the projects involve specialists in both AI and the relevant application domain, an approach that is critical to the successful translation of computational methodology into different fields. These interdisciplinary supervision arrangements ensure that graduates from the course will emerge as ‘bilingual’, confident in bridging communication and knowledge gaps in interdisciplinary research teams.
30 Business Impact • ISSUE 3 • 2024
Made with FlippingBook - Share PDF online