I have just finished reading this book.  It was referenced from The Future of the Professions.  

It was written by two storied experts

• Erik Brynjolfsson is a Professor and Senior Fellow at the Stanford Institute for Human-Centered AI (HAI).

• Andrew McAfee is a principal research scientist at the MIT Sloan School of Management and co-founder of the MIT Initiative on the Digital Economy. He is a leading expert on the impact of digital technologies on business, economics, and society.

Written in 2013.  It is not as far sighted as Nick Bolstrom’s super intelligence.  It feels more near future than far future.  The difference in perspective is important.  If you look at a landscape over geological time you see continents move and mountains rise.  If you look at the same landscape over generations you see earthquakes and and eruptions.  

Setting the historical perspective

The first half of the book Lays out how fast things are changing and makes the case for why the pace of change will accelerate.  One anecdote that struck me was when they talked about electrification of factories.  When steam powered the factory, there was one power source being mechanically distributed to work centers.  Work that required lots of power had to be close to the power source due to the stresses of transmission.  When electrification hit, factory layouts did not change for a generation.  They simply exchanged the steam engine with a big electric motor.  Only when a set of factory managers retired did the factory layouts follow the flow of the workpiece. 

Laying out the prize and the problem

In the second part of the book, the authors explored the potential bounty of the ai revolution.

They did then go on to explore the concentration of wealth that may also result from the same revolution, both it's consequences and it's remedies.  

Anecdote from this section is that while humanity ceded the game of chess to deep blue in 1997, teams of humans and computers do better than computers alone in chess.  

The large scale framing of a problem; to know where to focus; was seen as a particularly human advantage in 2013.  I am less certain of this in 2025.  I have been astounded at the capability of LLM’s to understand complex problems and structure and decompose responses.

Possible Remedies

According to the authors, part of the remedy is in education.  They pointed out how different the outcomes have been for different levels of educational attainment over the past 30 years. 

They pointed out that our current educational system produces people that would fit well into the civil service of the British empire. 

I really wish this theme had been explored more fully, because it is the first time that I have seen authors think about societies as a system with objectives and components , both human and machine.  

They seem very critical of current  university education.  Having been part of that system I think I agree, although, it was not clear how they thought it should be improved.  Read more, study more, write more is a fine mantra for a student, but is unhelpful as a guide for a university.

They clearly believe that massive online open courses (MOOC’s) are part of the solution.  

I was fascinated that they really backed Montessori as an educational model: self guided, curiosity driven with a healthy disrespect for rules. They thought this was the best training for the large framing of problems.  They noted that a disproportionate number of the leaders of technology companies come from a Montesorri system.  

Policy Recommendations

They have the courage to make some policy recommendations across education, science and innovation, tax policy.  They pointed out that while there are more conservative and more liberal economists, their doctrines mostly overlap, and economic growth will soften whatever the blow may be. 

They look further into the future to make some longer term recommendations regarding the world of work and the value of the structure of work in keeping societies stable, 

They did not explore the economic model when a large proportion of society can no longer sell their time for either their mechanical or intellectual efforts.  They seemed to show us the trajectory we are on and then briefly examined a couple of proposals for providing basic needs.  Clearly more thinking is needed in this area.