Bodies in Systems: Algorithmic Promises of Cure

Artificial intelligence (AI) in health care, like in many other areas, raises major hopes: quicker diagnoses, more precise treatments, and administrative easing, for example in patient documentation.

Instead of conventional tissue scans, the artist duo kennedy+swan fed an AI system specialized in cancer detection scans of watercolors on glass that imitate microscopic specimen. The system provided convincing diagnoses of the abstract works, and the aesthetic intervention proves that many of the systems tend to make plausible diagnoses rather than admit uncertainty. Likewise, current debates in AI research also revolve around the question of the systems’ reliability in case of incomplete or imprecise datasets.

Add to this that AI applications are often based on data sets that reflect the discrimination of already marginalized groups and social power relations. These inequalities result in, among other things, unequal access to medical care. False diagnoses can be difficult to discern for both young doctors and software developers. The solution often is an ex post facto elimination of bias, disguising it as a technical problem.

The panel discussion will look at the topic from scientific, societal, and artistic perspectives. AI obviously transforms countless aspects of society, and we ask: How do these technologies change our understanding of responsibility and trust? Which interdisciplinary approaches enable us to figure out how to deal with AI’s transformation potential?

Important Information
As spaces are limited, please feel free to register in advance. Your reserved seat will be held until 10 minutes before the event begins. If you cannot attend, kindly cancel your ticket so another guest can join.
Please note: late arrivals may have their spot reallocated.

Host or Publisher Berlin Science Week