Leopoldina Warns of AI's Deceptive Potential

Germany's National Academy of Sciences highlights risks and benefits of generative AI in a new discussion paper.

The Leopoldina emphasizes the need for a realistic view of AI's capabilities and limitations, cautioning against attributing human-like qualities to AI systems.
The discussion paper highlights the potential for AI misuse, particularly in generating propaganda or deepfakes due to its ability to create realistic texts, images, and videos.
Generative AI's reliance on its training data and development goals poses challenges in control, as these factors often escape institutional or normative oversight.
While acknowledging AI's potential to support everyday tasks, the paper stresses the importance of transparency, especially in sensitive areas like the legal field.
Initial efforts are underway to address AI's opacity and bias, but the Leopoldina warns against setting unrealistic expectations for these solutions.