The recent new wave of AI tools has been a wake-up call for many, making people marvel at the massive advances the technology has made and its potential to contribute across all domains of work, innovation, productivity and skills. But there are worries about the disruptions it will bring and the risks – even existential ones – that it can pose to society. In this session, after opening remarks from the OECD and Germany’s Federal Ministry of Labour and Social Affairs, Professor Stuart Russell will present the latest developments in AI, what they mean for humans and what needs to be done to make AI beneficial for humanity.

In this session, OpenAI will give a short demo of what GPT-4 is able to do.

AI chatbots has been front and centre in the news for past couple of months. They can quickly and cheaply produce well written essays, answer costumer support questions, and even pass exams. But AI chatbots can also incorrectly answer basic questions or generate hate speech and misinformation. This session will start with a keynote speech from Professor Ajay Agrawal on the implications of AI chatbots for job automation, work and productivity. To understand the policy implications and promote the adoption of trustworthy AI, Stephanie Ifayemi (Partnership on AI) will moderate a panel discussion with: • Emilija Stojmenova Duh (Minister of Digital Transformation, Slovenia); • Lucilla Sioli (Director for Artificial Intelligence and Digital Industry, European Commission); • Zoë Baird (Senior Counsellor to Secretary Gina Raimondo, U.S. Department of Commerce).

What actually happens in businesses and to workers when AI is introduced to the workplace? Stefano Scarpetta (Director of the Employment, Labour and Social Affairs Directorate, OECD) will present new OECD work, based on a survey of over 7,000 employers and workers and over 100 in-depth case studies. Businesses and workers will then share their own experiences of AI adoption while policymakers will discuss the measures they are taking to adapt. The impact of AI on the workplace: Main findings from the OECD AI surveys of employers and workers The impact of AI on the workplace: Evidence from OECD case studies of AI implementation

AI brings benefits but also real risks like bias and discrimination, the polarisation of opinions, privacy infringement, and widespread surveillance in some countries. Some of these risks are already materialising. Around the world, governments and AI actors have been converging around the use of risk-based approaches and impact assessments to help govern AI. But the frameworks and standards developed by different actors will need to interoperate as much as possible. Otherwise, achieving trustworthy AI will be not only more complex and costly, but also less effective and enforceable. The session will also present the Catalogue of AI Tools and Metrics a one-stop shop for AI actors to share and find approaches, mechanisms, and practices to prevent and mitigate risks and implement trustworthy AI.