windy day on the ferry
the usual suspects
dancing in the street
making jewelry
hi, i'm
Aarushi Lakhi
cs + design + entrepreneurship @ ut austin
CS student at UT Austin. I'm doing AI safety research, leading Austin AI Alignment, and I started a nonprofit at 12 that has funded 69 eye surgeries.
I've been thinking a lot about AI control and scalable oversight lately. I think the decisions we make about how to build and govern AI systems are some of the most consequential ones our generation will face, and I want to be in the room when they're made.
currently
- president of Austin AI Alignment
- researching how LLM agents follow rules when they use tools, through EleutherAI's SOAR fellowship
- taking Scott Aaronson's AI alignment theory course
- reading: When Breath Becomes Air & Peace Is Every Step
