PRISM 2026: Final Poster Showcase
Twelve teams from the PRISM AI Safety Research Fellowship's 2026 present 16 weeks of work in a live poster session, run like a conference poster hall.
How it works: a 10 minute intro, then twelve parallel rooms, one per team. Drop into whichever projects interest you and move between them at your own pace. Two hours in total, come and go as you like.
Preview first: each team's presentation video goes up on our YouTube channel in early October, so you can decide which rooms to visit.
The 12 projects:
How AI Labs Redefine Safety
Mentor: Aryan Agarwal
Fellows: Caitlin Mah, Lisa V, Rachita Mehta, Sakshi Chaubey, Yejin Kim
Do Rules and Concepts Hold Across Languages?
Mentor: Aviram Berg
Fellows: Anu Adesina, Ifeoma Okoh, Janhavi Khindkar, Maria Lomaeva, Ayesha Imran
Decision Update Consistency in LLMs
Mentor: Ayodeji Ibitoye
Fellows: Adepoju Jeremiah Moyondafoluwa, Alexey Kuntsevich, Bouthayna Mesraf
Estimating Harm and Exposure for AI Harm Monitoring
Mentor: Branwen Owen
Fellows: Danielle Dalisay, Irum Younis Khan, Sarah Quayyum, Sharifah Omar
Red-Teaming Protein Foundation Models
Mentors: Jeyashree Krishnan and Ajay Rangarajan
Fellows: Aayush Gandhi, Vrinda Nair, Girardy Momplaisir
Grounding Safe-by-Design AI
Mentor: Jobst Heitzig
Fellows: Amna Jamal, Anidipta, Gagana M D, Souhaila Djaffal (Past: Oula, Ayman Tarig, Ethar Ammar Mohamed Hamid)
Interpretability for Scientific Causal Reasoning
Mentor: Manjari Narayan
Fellows: Afnan Alabdulwahab, Laurent Vong, Nice Cailie Ineza, Sanskriti Shindadkar
Multilingual Safety Evals
Mentor: Naiyarah Hussain
Fellows: Abdullah Ibne Hanif Arean, Fathima Raihaan Ihsan, Fatma Egal, Victor Olufemi
Synchronous Monitoring for AI Control
Mentor: Preeti Ravindra
Fellows: Henry Chen, Isabella Luong, Jafar Isbarov, Kemunto Ochwang'i
Interpreting Personalized Reward Model Bases
Mentor: Rachel Freedman
Fellows: Harsh Sharma, Hassan Suliman, Ifesi Onubogu, Mohamed Eldagla, Prerana Rane
Preference Drift in Language Models
Mentor: Rubi Hudson
Fellows: Bukhaar Ali Mahamud Mahamed, Calissa Man, Hilary Torn, Xianglin Yang
Trust Calibration in Healthcare AI
Mentor: Zeenath Reza Khan
Fellows: Favour Chisom Akabogu, Kumari Neha Priya, Rokhaya Diagne, Tustee Mazumdar
Program Support Staff & Volunteers: Ayesha Imran, Sakshi Chaubey, Gagana M D, Amna Jamal, Lara Nguyen, Franziska Heuschkel, Gagandeep Kaur
Open to the public. Come if you work in AI safety and want to meet researchers building their first body of work in the field, or if you're weighing a future PRISM application and want to hear about the program from fellows themselves.