What AI jobs pay, skills matter most, and an AI bubble

A Note from Malcolm

From tulips to Teslas. 

Are we in an AI bubble, or are the valuations of AI bellwethers justified?

In the past three months the combined market caps of NVIDIA and Tesla have increased nearly $1.2 trillion (and that includes last week’s combined retraction of $220 billion). Such rapid valuation rises, combined with volatility, have all the hallmarks of a speculative bubble.  

I had a front-row seat for the ”time of the great happiness” (better known as the Internet bubble). Remember Webvan, Pets.com or eToys? A tremendous amount of cash - and a lot of reputations - went up in a flash. However, during that time Amazon was also called out for being wildly overvalued. Many argued the e-tailer should be valued like any retailer, such as Walmart or Sears. 25 years later Amazon is 71x its “inflated” price of July 1999. Walmart is up 4x and Sears, long bankrupt, now only has 10 remaining stores.

What’s the lesson? The Street can get the company names and market timing wrong, but is usually right about fundamental tech shifts. From railroads to the Internet, Amara’s law has held: we tend to overestimate the impact of a new technology in the short term and underestimate it in the long term. Will AI hold to this pattern or are we experiencing something new?   

With that lens, here are some highlights from the past week at the intersection of enterprise AI and work: