Hello everyone,
I am Wajahat. I did my Bachelors in Electrical Engineering, during and after that time I did a few internships at research centers and found myself more gravitated towards studying AI so I did a Masters in AI. My research area was Multimodal learning specifically application of multimodal models such as CLIP in image restoration and medical image generation (PET from MRI) and its interpretability. During that time I also learned about mechanistic interpretability and then found my growing interest towards AI safety.
Although I think of myself... (read more)
Thank you for sharing this, I never even considered that such a thing can be done and I think the AI safety should go both ways (safety of AI from Humans and safety of Humans from AI).