The next Shocklab Seminar will be held online on Wednesday, 1April 2026 from 16h00 - 17h00 (UTC+2).

Please join us in the Mathematics building for the second talk in our series by the Cooperative AI Research Fellows. Akash Kundu from the Heritage Institute of Technology will present “Similarity as a Signal: Do AI Agents Cooperate More When They Know They're Alike?"

Title: Similarity as a Signal: Do AI Agents Cooperate More When They Know They're Alike?
Speaker: Akash Jundu
Date: Wednesday, 1 April 2026
Time: 16:00-17:00 (GMT +2)
Venue: Room M304, UCT Department of Mathematics and Applied Mathematics Mathematics Building, 7 University Ave N, Rondebosch, Cape Town, 7700 
Virtual Meeting Link:  https://bit.ly/shocklab-seminars

 
Abstract: 
The Nash equilibrium for the Prisoner's Dilemma is to defect. Always. But here's a thought: what if you knew the coplayer across from you thought about the world the same way you do? Would you still defect? That's the question we're trying to answer — except instead of people, we're using AI agents. I'll share some early findings from ongoing experiments, a few things that surprised us, and plenty of open questions we haven't resolved yet. Thoughts and feedback very welcome
Bio:
Akash Kundu is a final-year Computer Science undergraduate and Cooperative AI Research Fellow with experience in technical AI Safety, focusing on evaluating and stress-testing large language models. His work has uncovered behavioural failures across a range of dimensions — including dark patterns, sycophancy, harmful reasoning, and multilingual vulnerabilities. He has co-authored research presented at ICLR, AAAI, and NeurIPS, and has collaborated with Apart Research, FAR AI, and Humane Intelligence on evaluation pipelines, adversarial prompting, and cross-cultural red-teaming.
 

Housekeeping: