Thinkers and practitioners from a wide variety of fields—from the tech sector, to medicine, to politics, to the arts—came together at the 2021 SNF Nostos Conference on the theme of Humanity and Artificial Intelligence. Behind the incredibly complex topic of artificial intelligence are simple, human essential questions: What sort of world do we want to live in? And how do we get there?
The concept of artificial intelligence has been on humanity’s mind for thousands of years, and approaches have differed profoundly across cultural contexts, which participants explored in a panel on Narratives and Visions. “The first narrative for an intelligent machine was in the Iliad….People have been thinking of creating objects that are as intelligent and can interact with humans for millennia,” said Kanta Dihal, Senior Research Fellow at the University of Cambridge Leverhulme Centre for the Future of Intelligence.
“AI is not one thing, it is multiple technologies through a system of power. The question is how we can design a system were power is shared,” said Anasuya Sengupta, Founder & Co-Director of Whose Knowledge, in a panel focused on envisioning AI Futures Worth Wanting. Power dynamics—how they influence the creation of AI technologies and how AI technologies help shift or reinforce them—were a consistent undercurrent across a number of different discussions.
“We can’t ask ‘Is it good?’ We have to ask ‘Good for whom?’…. We are making tradeoffs all the time,” said Google AI Software Engineer, Software Architect, and Designer Blaise Aguera Y Arcas.
Even when AI is being used to deliver more widely shared benefits rather than harnessed for the benefit of a few, the outcomes may not be equal or fair. In the medical field, Dr. Barry Coller, Vice President for Medical Affairs at The Rockefeller University, commented in a panel on Meta-Patterns, “We have to deal with the fact that something may be fantastic for general decision-making but can be a catastrophe for the individual.”
In the arts, William Kentridge observed that algorithms are very good at optimizing, but that the “less good idea” is often what produces the best result. The discussion panels of the conference were interspersed with short performances related to artificial intelligence by artists including Kentridge, adding an additional avenue for inquiry and exploration.
As AI has the potential to fundamentally reshape our institutions, our social structures, and our ways of living, conference participants probed how those same structures, habits, and unquestioned assumptions have helped shape the form that AI has taken. Zeid Ra’ad Al Hussein, President and CEO of the International Peace Institute, discussed how AI should be deployed during a panel on Discourse and Democracy. “We are now reaching some sort of existential moment and we need to change the competitive instinct, the commercial desire for profit into something that will allow our species to continue and not to destroy itself.”
“AI is trained by data. Data carries values. Values carry judgements. And therefore it embodies certain ethical perspectives on what is right, what is fair, and it embodies certain biases, conscious or unconscious, of a society,” said Nicolas Economou, Chair of the Science, Law, and Society Initiative at the Future Society and Principal Coordinator of the Athens Roundtable on AI and the Rule of Law in a panel on Connection, Competition and Cooperation.
The ethical challenges of developing new AI technologies concern not only the humans who interact with them, but also the technology itself. “I think on the way to building a machine that is conscious, that could think like a person,” said author Ted Chiang, “you will inevitably build a machine which is capable of suffering and then you will create a class of entity which is almost infinitely reproducible, which can experience suffering. Right now we have human suffering and we have animal suffering. I don’t think we need to create an additional category of entity which is capable of suffering.”
On the other hand, several speakers expressed optimism that seeing our own faults manifested and reflected back to us in artificial intelligence could be the impetus humanity needs to address our prejudices. “Is AI biased? It is like looking in the mirror and not liking what you see,” said Blaise Aguera Y Arcas.
However the relationship between humanity and AI evolves in the long term, a sense that the this evolution is already underway—and perhaps further along than we can appreciate—was widely shared. The first day of the conference closed with perspectives from the people who will be most affected by accelerating technological change: young people. Five students fielded questions, then turned the tables to ask questions of the adults in the room.
The discussion with the students struck an optimistic tone on the future of technology, affirming that real human connection is irreplaceable, and that younger generations are conscious of both the upsides and the downsides of using social media. Because perspectives like these, from the young people who will experience accelerating technological change more than anyone, are too often missing from the conversation, the Stavros Niarchos Foundation (SNF) announced the launch of a brand-new opportunity for them to share their perspectives. The SNF 25th Anniversary Short Film Challenge invites people ages 14 to 21 to share their ideas about how technology can be a force for good and win a chance to take their story to the next level with professional creators.
In his closing remarks, SNF Co-President Andreas Dracopoulos issued a collective call to embrace and engage with AI, and to direct our efforts toward shaping AI so that it better serves humanity. “We all need, as much as we can, to do all we can to help improve education, healthcare, justice, decency, compassion—to bring back hope, for every human soul in this world.”
Conference minutes and summary provided by iMEdD.