As a Machine Learning Engineer with two master's degrees in AI, I have been privileged to observe and participate in the ongoing AI revolution from a front-line perspective. My hands-on experiences with foundation models have solidified my belief that the boundaries of AI are inexhaustible.

I envision a future world transformed by AI, where humanity thrives on a global scale. In this world, humans are free to pursue their passions, while advanced AI agents construct their houses and provide healthcare. However, this future is not without its challenges. The accelerating advancement and adoption of AI raises complex issues surrounding ethics and control. Moreover, the impact of a misbehaving system grows as AI agents become more deeply ingrained in our lives.

This is why I believe in the importance of AI safety. This domain studies problems like machine alignment and mechanistic interpretability to safeguard the adoption of advanced AI. By prioritizing AI safety, I believe we can build a world where we thrive alongside intelligent systems. We have one chance to do it right, and we must act now!

I am a proud member of the Effective Altruism community. I joined the AI Safety Fellowship and in December 2022, I participated in the EAGxBerkeley conference.

For those interested in the possibilities and risks of advanced AI, I recommend reading Our Final Invention by James Barrat and Superintelligence by Nick Bostrom.