Just posting to express my appreciation for the rich discussion. I see two broad topics emerging that seem worthy of systematic exploration:
What does a world look like in which AI is accelerating the productivity of a team of knowledge workers by 2x? 10x? 50x? In each scenario, how is the team interacting with the AIs, what capabilities would the AIs need, what strengths would the person need? How do junior and senior team members fit into this transition? For what sorts of work would this work well / poorly?
Validate this model against current practice, e.g. the ratio of junior vs. senior staff in effective organizations and how work is distributed across seniority.
How does this play out specifically for AI R&D?
Revisiting the questions from item 1.
How does increased R&D team productivity affect progress: to what extent is compute a bottleneck, how could the R&D organization adjust activities in response to reduced cost of labor relative to compute, does this open an opportunity to explore more points in the architecture space, etc.
(This is just a very brief sketch of the questions to be explored.)
I’m planning to review the entire discussion here and try to distill it into an early exploration of these questions, which I’ll then post, probably later this month.
Just posting to express my appreciation for the rich discussion. I see two broad topics emerging that seem worthy of systematic exploration:
What does a world look like in which AI is accelerating the productivity of a team of knowledge workers by 2x? 10x? 50x? In each scenario, how is the team interacting with the AIs, what capabilities would the AIs need, what strengths would the person need? How do junior and senior team members fit into this transition? For what sorts of work would this work well / poorly?
Validate this model against current practice, e.g. the ratio of junior vs. senior staff in effective organizations and how work is distributed across seniority.
How does this play out specifically for AI R&D?
Revisiting the questions from item 1.
How does increased R&D team productivity affect progress: to what extent is compute a bottleneck, how could the R&D organization adjust activities in response to reduced cost of labor relative to compute, does this open an opportunity to explore more points in the architecture space, etc.
(This is just a very brief sketch of the questions to be explored.)
I’m planning to review the entire discussion here and try to distill it into an early exploration of these questions, which I’ll then post, probably later this month.