If his example is Instagram then the lesson is that you didn't use to need many employees to get rich via venture capital and buy out, which are the main goals

If his example is Instagram then the lesson is that you didn't use to need many employees to get rich via venture capital and buy out, which are the main goals of many tech founders. Instagram and tech in the early 2010s is not a company or sector that represents the wider economy. Instagram in 2012 had 30 million active users and zero revenue. That's $0 value earned per employee; these days it's just under $2 million per employee. The scale is vastly different and they couldn't still be running with so few employees. And let's not forget that Instagram was bought as it was seen as a threat to facebook. That it was seen as worth a billion dollars is exactly the problem that many people are referring to about creating overnight billionaires but overvaluing digital companies, which would be worth much less if they were brick & mortar companies. That said, there are a lot of great tools out that that allow people to run companies with off the shelf software and no code applications, which I think have the potential to make a bigger impact on software development than AI. Companies using new technology to become more efficient is nothing new, it's as old as work itself. The steam engine automated a lot of work but still required workers, just like automation tools require another company's employees to build. Growth creates more employment and automation moves those roles elsewhere. The problem is more that those employees that the enriched founder does need working for them are disrespected and paid badly as they are seen as just cogs in the machine who should be happy for the breadcrumbs tossed to them. Doesn't matter how many employees you have, if the founder is getting rich they can afford to pay their staff well.