Hey all, I hope writing my introduction in a quick take was a sound idea.
I am 19, from Iraq, a third world country and I am an AI and robotics engineering student. I am also a co-founder for one of the biggest annual events on AI in Iraq. I recently discovered the fields of AI safety and mechanistic interpretability and I am very interested. I see my generation of students to be very interested in applying AI meanwhile I am going through a lot of thinking on what is healthy and secure AI. I mean we are using a system that is inherently a black box and people are believing whatever it outputs. I don’t trust that big AI corporations have safety as their primary goal.
Other than AI, I am well read in philosophy and I am highly interested in making sense of the world and pursuing truth. I have shifted my beliefs many times and I keep unpeeling layers deeper into my understanding of reality and myself.
I have navigated many existential crisis experiences and also some de-personalization experiences.
I would love to connect to anyone is interested in science, engineering, AI, mech interp, and philosophy or someone with more experience who can give me tips especially on the mech interp domain.
Hi! Chris Olah was head of interpretability at OpenAI and now at Anthropic, he is probably an important person. @Neel Nanda might be the guru of interpretability on this forum (his guide to starting out).
Hey all, I hope writing my introduction in a quick take was a sound idea.
I am 19, from Iraq, a third world country and I am an AI and robotics engineering student. I am also a co-founder for one of the biggest annual events on AI in Iraq. I recently discovered the fields of AI safety and mechanistic interpretability and I am very interested. I see my generation of students to be very interested in applying AI meanwhile I am going through a lot of thinking on what is healthy and secure AI. I mean we are using a system that is inherently a black box and people are believing whatever it outputs. I don’t trust that big AI corporations have safety as their primary goal.
Other than AI, I am well read in philosophy and I am highly interested in making sense of the world and pursuing truth. I have shifted my beliefs many times and I keep unpeeling layers deeper into my understanding of reality and myself.
I have navigated many existential crisis experiences and also some de-personalization experiences.
I would love to connect to anyone is interested in science, engineering, AI, mech interp, and philosophy or someone with more experience who can give me tips especially on the mech interp domain.
Hi! Chris Olah was head of interpretability at OpenAI and now at Anthropic, he is probably an important person. @Neel Nanda might be the guru of interpretability on this forum (his guide to starting out).