The vast majority of AI use in 5 years from now will be on things we don’t even do today. In most areas of work, we actually haven’t reached peak demand. In fact we’re very far from it. We haven’t reached peak demand of getting great healthcare services, coding amazing apps, fixing bugs, building features customers want, protecting companies against risk, delivering better sales experiences, creating relevant ad campaigns, and so on. In all of these areas, and 1,000’s of other economically valuable activities, we only happen to do the level of work that we do because it’s all a company can afford to do. There are endless tasks in a company that they would do 10X more of if they could afford to; and there are endless companies that would love to do certain activities for the first time if they could afford to. AI makes both possible. As a result of this, it also means most of the dire predictions on jobs will be wrong. What will ultimately happen with AI is some companies will use AI to save money and do the same; other companies will use AI to do far more. Over time the companies that use AI to do far more will compete better, and that will eventually change customer expectations until others in the market have to match. We’ll wake up in 5 years from now not quite knowing how we got to where we got, but where everyone just works at a higher level of abstraction than before, but getting more done, moving projects along faster, delivering better customer experiences, and so on.

Insightful interview with @JonasAndrulis. This part is particularly spot on. The EU’s AI Gigafactory program is another misguided initiative that’s bound to tie up scarce resources without creating strategic value.

Es braucht nur eine Seite, um mit der Mär von der Regulierungssupermacht Europa & dem angeblichen Wettbewerbsvorteil durch den Brussels Effect aufzuräumen. Gute Wochenendlektüre von @PatrickBernau.