Inaugural Lecture: Human and Artificial Intelligence

Join us at the ESCP Berlin Campus for Professor Florian Artinger’s inaugural lecture on Human and Artificial Intelligence

Human cognition achieves remarkable efficiency with limited resources: for instance, the brain operates on about 20 watts of power, roughly the energy consumption of a computer monitor in sleep mode. At the same time, developers hope to build machines that think and learn like humans. In pursuing this goal, AI models have become ever more complex, requiring informational and computational resources that vastly exceed those used by humans. How can people achieve so much with so little?

What you'll learn

• Human cognition is frugal: People often make effective decisions using only a small amount of relevant information instead of all available data.

• Recency as a predictor: The time since an event last occurred can be a powerful predictor of future outcomes, even when used alone.

• When simple beats complex : Under certain real-world conditions, relying only on recency is ecologically rational and can perform as well as or better than more complex predictive models.

• Evidence across industries: Studies across 58 real-world tasks in finance, healthcare, and consumer purchasing show that recency-based models outperformed or matched more complex models in 62% of cases.

• Building better AI: Insights from human cognition can help create machine learning algorithms that are more accurate, efficient, and transparent.

• Business and regulatory impact: Explainable AI models, such as Schufa's credit score based on 12 variables, demonstrate how transparency can meet regulatory requirements while improving human–AI decision-making.