The Sylvan artificial intelligence debate considered the following motion:
This House believes that AI is more likely to reduce than increase happiness.
The debate took place on Monday, 2nd March. It was proposed by Paul Byford and opposed by Lynn Jiang.
The proposition supporting the view that AI is more likely to reduce happiness
The proposer covered a range of areas where AI is likely to have an impact on human society. While there is a large amount of uncertainty about what AI will become, humans are the driving force behind it. He argued that humans use tech for a range of things, from making war, controlling others and generating profits (by companies). He outlined potential military applications of AI, which would certainly not lead to happiness. In addition, the Chinese government is using AI for facial recognition and a social capital system – the type of applications which, if used here, could lead to data insecurity and a reduction in freedom and spontaneity in society. Automation of customer service by companies to reduce costs takes away little human interactions which, for isolated people, could have a large impact. He also stated that AI would lead to increased inequality, with profits skewed to those who could leverage AI.
The opposition against the motion
The opposer addressed a number of those points, argued that AI destroys class barriers and shared a range of tangible benefits generated by AI. For example, recommendations made by online services based on past order history saving time for consumers. Crucially, sharing of data for AI is voluntary. AI will provide enhanced outcomes without the irrational biases of humans – for instance in medical diagnoses. It can also be used to help individuals develop their own uniqueness, e.g. via digital tools to support artistic endeavours. Overall, she argued that AI will create benefits that will provide a better standard of living for ordinary people, freeing up their time to pursue leisure activities of interest to them.
Floor speeches picked up on a number of threads, and in particular one in which the definition of happiness was explored. Maslow’s hierarchy of needs was put forth as one model, though a number of speakers contended it is too simplistic (nevertheless, many referred to it). The threat of ‘general’ AI whereby AI systems could create their own intelligence outside of human control, and the point of AI being more intelligent than humans (even to the point of potentially out-debating the best debaters!) was raised. However, a useful counter point was that chess is still enjoyed by millions, even though the best players cannot beat a computer, and in go, the game of human players has been improved through play against AI. The impact on workers and the potential for job insecurity in the service sector could have a major effect. A number of speakers discussed how the use and control of AI itself would play a major role in whether it would reduce or improve happiness overall. Multiple speakers touched upon the impact of new technology on those less able to adapt to it, often the older generations.
Result: in the final vote, the artificial intelligence debate motion carried
The Sylvans concluded through the artificial intelligence debate that AI is more likely to reduce happiness than increase it.
See information on other Sylvan debates here.