This post is a great crash course in where we are with AI, how we got here, and what the future of Artificial General Intelligence looks like. 🔮
I generally think we’re a ways off. There are some really impressive examples of AI (like the standard AlphaGo and AlphaZero examples), but they also implode if the parameters of the task change or are asked to do something “off script”. They also require a lot of compute power. ⚡️⚡️⚡️
Our current AI boom is fueled by exploding data and compute power. It is turning older ideas into reality. I wonder if AI works on a weird cycle where breakthrough ideas occur and then stagnate until the tech catches up; there is an explosion, fears and feats accelerate, research booms and new ideas surface; is the next step another stagnation as we wait for the next tech catch up?