Part IV. Frontiers
IN THIS LESSON
This is the final lecture in the Language Models and Intelligent Agentic Systems course, run by Meridian Cambridge in collaboration with the Cambridge Centre for Data Driven Discovery (C2D3).
This lecture discusses forecasting AI progress over the next 5-10 years. We talk about historical trends in both increasing compute used to train models and algorithmic advancements in how models are trained. In parallel, we consider recent improvements in the time-scale of economically valuable work models are capable of performing. Finally, we discuss the prospect of AI being used to automate AI R&D.