How robots make us smarter
The first global standards for ethics in artificial intelligence show humans can better understand intelligence itself.
Did humanity just become genuinely smarter about artificial intelligence?
That could be the case with a 39-page set of recommendations released last week by the United Nations.
The document, a result of three years of work by hundreds of experts, was endorsed by 193 countries. For the first time on a global scale, it lays out universal values for the ethics needed to ensure that AI-driven technologies – from facial surveillance to driverless cars – “deliver for good.” Those values include transparency, social inclusion, accountability, and “explainability.”
“These new technologies must help us address the major challenges in our world today, such as increased inequalities and the environmental crisis, and not deepen them,” said Gabriela Ramos, an assistant director-general at the United Nations Educational, Scientific and Cultural Organization.
The new UNESCO standard is seen only as an “incentive” for nations to build ethics into their AI systems. Yet it adds to other recent efforts to bring a higher order of thinking to the use of computer algorithms now reaching a scale and speed not even imagined in the best of science fiction going back to Mary Shelley’s 1818 novel “Frankenstein.”
One of the most active efforts at setting AI standards is the Global Partnership on Artificial Intelligence. This coalition of 20 countries, set up in 2020 by France and Canada, aims to ensure that advanced technologies promote the values of democracy and equality.
The ethical challenges inherent in AI – especially robotic military weapons and mass surveillance – are many. The UNESCO report, for example, says the ultimate responsibility for AI technologies must lie with humans and that even the most intelligent devices “should not be given legal personality themselves.”
The challenges posed by machines able to learn and adapt are in turn forcing humans to learn and adapt. This helps expand human intelligence, forces a higher concept of consciousness, and puts a focus on the source of intelligence itself, beyond material origins. As more qualities of human thinking are built into AI, such as trust and empathy, the more the designers and regulators of AI must stretch their own thinking.
The UNESCO report is a good example of that progress. Humanity can be smarter, in a very tangible way, and “deliver for good.”