ADM+S Partner Investigator invited to speak about responsible AI at the United Nations
Author Natalie Campbell
Date 29 February 2024
On 7 February 2024, ADM+S Partner Investigator Prof Julia Stoyanovich from New York University was invited to speak on responsible artificial intelligence (AI) at the United Nation’s 62nd session of the Commission for Social Development (CSocD62).
The Commission took place from 5 to 14 February 2024 at the United Nations Headquarters in New York and focused on ‘fostering social development and social justice through social policies to accelerate progress on the implementation of the 2030 Agenda for Sustainable Development and to achieve the overarching goal of poverty eradication’.
Prof Stoyanovich was one of six panellists invited to discuss this meeting’s emerging issue, ‘the Influence of Digital Transformation on Inclusive Growth and Development: A Path to Achieving Social Justice.’
Defining AI as a system in which algorithms use data to make decisions on our behalf, or help humans make decisions, Prof Stoyanovich provided examples of successful AI implementation, where technology responds to a particular need to improve the status quo.
As an example, Prof Stoyanovich detailed how AI has significantly improved the efficiency of MRI’s and other technical medical services.
However, referring to a recent study about the use of AI in hiring processes and their amplification of biases, Prof Stoyanovich demonstrated how such systems can fail.
“You cannot outsource the work of being human,” she said.
“For safe use we must have another factor that contributes to their successful use and that is decision-maker readiness.
For example, a clinician or radiologist who is assisted by an AI, but understands they are ultimately responsible for treatment and diagnosis and knows when to trust AI predictions and when to challenge them.”
The panellists offered a range of perspectives on Responsible AI, AI Literacy and AI Governance to be considered by the commission, which is the advisory body responsible for the social development pillar of global development.
“We are technically ready to make [AI systems] in terms of data software and hardware, and we know how to validate them. These are some of the hallmarks of responsible AI.
“However, for safe use we must have another factor that contributes to their success, and that is decision-maker readiness,” Prof Stoyanovich explained to the Commission.
View the full hearing on UNDESA YouTube. (Prof Stoyanovich: 1:00h-1:15h)
Recent opinion piece from Prof Stoyanovich on regulating responsible AI.
