AI Manipulation Hackathon - Berlin
Are you worried about future AI systems using deception, strategic behavior, or psychological exploitation to achieve their goals at the expense of human values and intentions?
Join other fellow Berliners in participating in Apart Research's AI Manipulation Hackathon. Whether you want to work alone, remotely with another team or join others locally to build a team, we'll be hosting a jam site for you to work and collaborate from.
Location
The Teamwork coworking space in Wedding is graciously hosting us this weekend. The address is Müllerstraße 138D, 13353 Berlin. You can find detailed instructions for getting to Teamwork here.
Food & Drinks
Friday night we'll be ordering pizza for dinner. Arrive at 17:00 to get fueled up while meeting others participating in the hackathon. Throughout the rest of the weekend, we'll be provided some food, snacks and drinks to keep you going.
Schedule
Friday, Jan 9
17:00 - Doors open for networking & dinner.
18:00 - Opening Keynote: David G. Rand
19:00 - Hacking begins
Saturday, Jan 10
14:00 - HackTalk: Jan Batzner
19:00 - HackTalk: Kobi Hackenburg
Sunday, Jan 11
13:00 - Hacktalk: Lars Malmqvist
19:00 - HackTalk: Esben Kran
23:59 - All entries must be submitted
Prizes
The top teams will get:
$2,000 in cash prizes
The change to continue developing via Apart Research's Fellowship program
Guaranteed acceptance to present at the International Association for Safe & Ethical AI (IASEAI) workshop in Paris on February 26, 2026
Project Ideas
Projects can include:
Manipulation benchmarks that measure persuasive capabilities, deception, and strategic behavior with real ecological validity
Detection systems that identify sycophancy, reward hacking, sandbagging, and dark patterns in deployed AI systems
Real-world monitoring tools that analyze actual deployment data to catch manipulation in the wild
Evidence-based mitigations – MVPs demonstrating novel countermeasures with empirical backing
Multi-agent simulations exploring emergent manipulation dynamics and training processes that produce deceptive behavior
Pursue other empirical projects that advance our understanding of how AI systems manipulate and how we can stop them
Questions? Lost?
For questions or if you are lost, contact Trevor on Telegram @FastFedora.