“Large teams solve problems; small teams generate new problems to solve.” — Jeanne Brett and Dashun Wang, summarizing their analysis of 65 million papers, patents, and software products
Kind of an oblique critique of the top-down "Big Tech" approach to AI systems we've all experienced these past few years -- where all the innovation requires trillion-dollar budgets from gigantic incumbents, a human pregnancy cycle devoted to training a model on all the data in the known universe, and data centers the size of New Mexico.
DeepSeek briefly tried to remind us that a small-is-beautiful is still valid. But financial blitzscaling has no patience for small ideas.
Coincidentally, last month I gave a talk at the SU Summit Spain titled "Learning Faster Than Change". Great minds, yeah? 😅
Kind of an oblique critique of the top-down "Big Tech" approach to AI systems we've all experienced these past few years -- where all the innovation requires trillion-dollar budgets from gigantic incumbents, a human pregnancy cycle devoted to training a model on all the data in the known universe, and data centers the size of New Mexico.
DeepSeek briefly tried to remind us that a small-is-beautiful is still valid. But financial blitzscaling has no patience for small ideas.
Coincidentally, last month I gave a talk at the SU Summit Spain titled "Learning Faster Than Change". Great minds, yeah? 😅
Couldn't agree more - and so cool to hear about your session? Did they, by chance, record it? Would love to see it!
Internal use only I think. I'll find a way to share if they distribute it.