> I don’t know how Google could be saved but perhaps having grateful and happy employees that would just build things in their 20% time might have been a better

> I don’t know how Google could be saved but perhaps having grateful and happy employees that would just build things in their 20% time might have been a better strategy. I'm not sure 20%-time is a better strategy. At least it was not proven yet. For starters, which successful products were initially Google employees' 20% projects? Google Search was certainly not. Google Search Ads was not per the book In the Plex. The product went through multiple iterations, and a math whiz from u of waterloo conceived the idea of using second-price auction system. Google Maps was not, per the book Never Lost Again. It was developed by the Lars project initially at Where 2. GSuite was not. It was from the startup Writely. Google Adsense was not. I was from DoubleClick. Google Chrome was not. It was a strategic product that multiple people drove, including the current Google CEO. Google's self-driving car was not. Page personally drove the project by working with leaders like Thrun. On the other hand, breakthroughs like ChatGPT can't be from a 20% project. It requires full-time dedication of multiple world-class researchers and engineers to develop a series of models, to set up a massive infrastructure for training and data processing, to implement an efficient process for human input, and to invest many millions of dollars to buy GPU time and crawler time.