Skip to content
In PersonIn personPast event

Guest Talk: Automated Deepfake Detection with AI

This event has ended

This listing is kept as an archive. Explore upcoming events to find your next one.

See upcoming events
About this event

What to expect

Understanding the difference between real and fake content is becoming increasingly critical as generative AI models advance. This guest talk at the Hasso-Plattner-Institut tackles one of the most pressing challenges in media integrity: automated deepfake detection.

Jonas Bernhard from the AI Service Center Berlin-Brandenburg will introduce Clyravision, a sophisticated multi-agent system that uses AI to identify deepfakes through interpretable and transparent methods. Rather than treating detection as a black box, Clyravision combines multiple models and approaches to provide explainable results—essential for building trust in automated systems.

The talk explores both sides of the deepfake challenge: how generative models create increasingly realistic fake content, and how AI-based detection techniques can keep pace with these advances. You'll learn about the technical approaches that power Clyravision's multi-agent architecture and how different models work together to flag manipulated media.

Who Should Attend

This session is designed for anyone working with or interested in:

  • Trustworthy AI and responsible technology development
  • Media integrity and combating misinformation
  • Computer vision and generative model applications
  • AI safety and verification systems

Whether you're a researcher, developer, journalist, or simply concerned about the spread of synthetic media, this talk offers practical insights into one of AI's most challenging detection problems.

The event takes place on Tuesday, March 17, 2026, from 5:00–6:30 PM CET at the HPI campus in Potsdam. The venue is Campus Kitchen One, Building L, Room L-1.02. Note that photo, audio and video recordings will take place during the event.

Admission is free, though registration is required to secure your place. The talk runs for 90 minutes, allowing time for questions and discussion about the technical and ethical dimensions of deepfake detection.

Find more AI events in Berlin and explore other AI safety and ethics discussions in the Create With community.

[Register for this event](https://www.eventbrite.de/e/guest-talk-automatische-deepfake-erkennung-mit-ki-tickets-1981013405423?aff=oddtdtcreator&_cldee=cMOzVsIUoxoXmKrRHZD2CM5nPr7Ut-4wr4hVbidTn-aDcFMc8fI9_Qhh_o6jWr7XO4aG03OOMbfguBaDCYC0vg&recipientid=contact-ea4764959389

Running an AI event?

List it here, free.

Meetups, workshops and conferences go in front of thousands of builders after a quick review.

List your event