An artificial intelligence (AI) by IBM ‘took’ to the stage last week at the University of Cambridge to debate the dangers of AI.

The Robot called ‘Project Debator’ argued for both sides and ultimately managed to convince audience members that AI technology will do more good than harm.

AI technology will do more good than harm.

courtesy of https://www.newscientist.com/

There has been much talk since the 1950’s of whether morality and emotions could ever be programmed into thinking machines.

Currently we have narrow AI that solves specific problems and a currently unfathomable leap in AI technology would be needed for machines to make conscious choices. Project Debater’s words reflected this saying “AI will not be able to make a decision that is the morally correct one, because morality is unique to humans.” PD continued, “AI companies still have too little expertise on how to properly assess datasets and filter out bias. AI will take human bias and will fixate it for generations.”

“AI will not be able to make a decision that is the morally correct one, because morality is unique to humans.”

IBM’s AI called upon an application known as “speech by crowd” to produce its arguments. It analysed thousands of human submissions sent to it and then categorised these into key themes, removing any duplicates.

courtesy of https://www.newscientist.com/

Taking the opposition side, Project Debator began a pitch to advocate for net gains from AI, suggesting that AI will create new jobs and “bring a lot more efficiency to the workplace”. Unfortunately, these arguments lacked examples or specifics.

When put to an audience vote, the pro-AI side won a narrow victory with 51.22%. This result, however, might be more indicative of audience bias that the quality of the AI’s arguments.

The speech-by-crowd AI is being used more generally as a tool by IBM to collect feedback from large numbers of people. Use cases include helping governments seek public opinions about policies, or for companies seeking feedback from employees. IBM engineer Noam Slonim says “This technology can help to establish an interesting and effective communication channel between the decision-maker and the people that are going to be impacted by the decision,”

Leave a Reply