MozFest 2022: Can Artificial Intelligence Join the Fight Against Corruption? The role of AI in public spending

Governments around the world spend $13 trillion a year on contracts for goods, services and public works, but key information is not made public in over 97% of this spending, raising huge risks around value for money, corruption, and public health and safety. As the pandemic has further exposed the real human cost of this lack of transparency (i.e. in vaccine, PPE, and medical supply contracts), the role of public procurement monitors and corruption watchdogs has proven more critical—and life-saving—than ever.

But with limited capacity for manual monitoring of government spending and transparency efforts, more sophisticated tools to support daily procurement tracking are emerging, including artificial intelligence bots that are able to browse through public data and surface risks, irregularities, and red flags.

Given the inherent structural inequities present in many public procurement systems that enable corruption, how can we center Global Majority experts to ensure that new AI technology will not replicate the human (and often Western) biases built into them? What are the challenges and pitfalls of AI in public spending monitoring, and how might it expose corruption and help create more equitable social outcomes?

This panel discussion will bring together technologists, lawyers, data scientists, and civil society advocates from South Africa, Costa Rica, Paraguay, and Ukraine for a candid conversation on the role of AI in anti-corruption work, and it’s potential to transform the world’s largest marketplace: public procurement.

Panelists:

• Professor Sope Williams-Elegbe, Professor and Deputy Director, African Procurement Law Unit at Stellenbosch University
• Albert Sanchez-Graells, Professor of Economic Law, University of Bristol Law School
• Camila Salazar, Lead Data Analyst, Open Contracting Partnership
• Andriy Kucherenko, Product Manager at Dozorro, Ukraine 
• Fabrizio Scrollini, Executive Director, Latin American Open Data Initiative (ILDA)

The session will take place on 9 March, 1-2pm EST.