“Is emotion the last thing that separates human from machines?” With this quote Prof. Rosalind Picard opened her Tanner Lecturer on the 9th of June, hosted jointly by Reuben and Linacre as part of Reuben’s “AI for Good” series.
Prof. Picard began her career researching AI and smart machines. As this work progressed, she realised that the tools her team were building to help machines could also be used to help people overcome social and emotional difficulties. This realisation has led to decades of innovative research at the intersection of AI, emotions, and human behaviour at the MIT Media Lab.
As AI acquires increasing emotional intelligence skills, many people worry this represents an existential threat to humans. Prof. Picard reminds us that humans and machines remain ontologically different. Despite exhibiting increasing human-like social behaviours and functions, for instance a chatbot responding empathetically to someone’s sign of distress, a machine has no thoughts, feelings, or awareness. In fact, as research suggests, an identical empathetic response would be received more positively if delivered by a human than a machine. AI is ontologically inferior to its makers.
In contrast, she argues, what we humans beget is ontologically equal to us. Children and all living people carry transcendent value and worth. This fundamental worth is independent of physical or mental abilities such as age, gender, and ethnicity, achievements, or possessions. This quality of being human gives us the ability to practice compassion, humility, love, forgiveness, and many other virtues.
All people are equal – but are not treated equally. As technology makers, we can choose to use AI to exacerbate inequities, or to remedy them. Prof Picard urges us to embrace AI but acknowledge a higher standard of human values – one that acknowledges the equal worth of every living person. This includes “helping where there is need, not prestige”, and developing technology that is independent of functions or abilities and benefits a range of different people.
The many inventions that have come out of MIT’s Affective Computing Research Group certainly keep true to these values. A key innovation has been a smartwatch that anticipates seizures in people with epilepsy and sends an alert to a trusted contact. Another is a wristband that measures stress in individuals who are unable to speak, and helps others understand their feelings. Algorithms developed by her group have also allowed better forecasting of cognitive symptoms in people with dementia. Her work has also supported research and clinical practice in depression and post-traumatic stress, among many other areas.
Prof Picard’s research career is a statement about the potential of AI to improve human lives, not only by promoting health but also by enhancing values of justice and equality. May we all use the unique qualities that separate us from machines to build technology that helps people and societies to flourish.
Gabriela Pavarini is Nuffield Department of Public Health (NDPH) Intermediate Fellow at the Ethox Centre, and Research Fellow at the Wellcome Centre for Ethics and Humanities and Reuben College.
