You're pledging to donate if the project hits its minimum goal and gets approved. If not, your funds will be returned.
Travel support to attend EA Global: New York 2026 to connect with researchers and practitioners working on AI control, alignment, and interpretability.
EA Global will help accelerate my transition toward technical AI safety research by meeting researchers and practitioners working in AI control, alignment, evaluations and interpretability.
I recently published a solo-author paper at KDD 2026 on the reliability of probe-derived control directions for steering model behavior. I want to use the event to identify research collaborations, get feedback on my current research direction, and better understand where my technical background can contribute to AI safety.
I plan to engage as volunteer and network with fellow attendees and follow up on promising research connections afterward.
The funding would primarily cover the costs of attending EA Global from India:
- Flights: ~$750
- Accommodation( 4 nights): ~$580
- Approx Total: ~$1,330
I am independently exploring research areas in AI control and solving for gradual disempowerment.
My solo-author paper "Readable ≠ Steerable: Why Probe-Derived Context-Trust Directions Fail as Reliable RAG Control Surfaces" was accepted as an oral presentation at Secure and Trustworthy LLM workshop at KDD 2026. I have also worked on scalable expert-in-the-loop AI systems for healthcare and have 5 years of industry experience building production systems.
I am currently developing my research direction around AI control, interpretability, and reliable human oversight, and have recently completed BlueDot Impact's Technical AI Safety program.
Here is my google scholar: https://scholar.google.com/citations?user=SZtyqqgAAAAJ&hl=en
I will be unable to attend because of the travel cost. I would lose the opportunity to have in-person conversations with researchers and practitioners whose work is closely related to my research direction.
I would still continue my research but attending EA Global would increase my ability to find collaborators, receive feedback, and coordinate with the broader AI safety community.
I have not raised any funding.