Christopher Summerfield

Stunning advances in digital technology have given us a new wave of disarmingly human-like AI systems. The march of this new technology is set to upturn our economies, challenge our democracies, and refashion society in unpredictable ways. We can expect these AI systems to soon be making autonomous decisions on the user’s behalf, with transformative impact on everything we do. It is vital we understand how they work.

Can AI systems ‘think’, ‘know’ and ‘understand’? Could they manipulate or deceive you and, if so, what might they make you do? Whose interests do they ultimately represent? And when will they be able to move beyond words and take actions for themselves in the real world? Ultimately, can we look forward to a technological utopia, or are we in the process of writing ourselves out of history?

Christopher Summerfield is a Professor of Cognitive Neuroscience at Oxford and serves as the Research Director at the UK AI Security Institute (AISI), the UK’s pioneering state-backed organisation dedicated to ensuring that advanced AI technology is both safe and beneficial for society.

Established following the UK’s inaugural major AI safety summit in November 2022, the AISI aims to enhance global understanding of AI safety through rigorous research and evaluation of emerging AI systems and their associated risks, such as misinformation and mental health impacts. Collaborating with the wider research community, AI developers, and governments, the Institute not only seeks to shape the development of AI but also to influence global policymaking.