The Automation-to-Choice Cascade
Automation kills the routine and hands you the harder question: what next?
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 4
- Confidence
- 80%
Wolfram frames AI as the latest step in a long historical pattern of automation. Once an objective is defined, automation to achieve it can in principle be built — but the harder, enduring question is what to do next, and there are an infinite number of new things one could do. Historically, automating a big chunk of self-done work (agriculture, telephone switchboards) destroyed those jobs but opened many new categories, like the podcasting enabled by telecom automation. The pattern: new activities start as a matter of human choice, eventually get standardized, then automated, freeing humans to move to the next stage. The durable human contribution is deciding which slice of the infinite possible to pursue.
Origin
Wolfram generalizes from economic history — 150 years of US agriculture mechanization, the automation of telephone switchboard operators — and his own 40-year effort automating low-level programming, into a repeating cascade from routine work to human choice.
Core principles
- 01Once an objective is defined, automation can eventually be built to achieve it.
- 02Automating a big category of work opens up many new categories of work.
- 03The infinite set of things that could be done means the scarce act is choosing which to do.
- 04Choosing what to do next is the durable human role; routine work standardizes then gets automated.
How to run it
- 1
Spot the routine portion
Identify the part of your work that is just turning the crank — routine effort that doesn't require much thought.
Watch out Betting a career on the crank-turning skill is risky; it's first in line to be automated.
- 2
Expect it to be automated
Recognize that once work becomes routine and standardized, it can and will be automated.
Pro tip Machine-learning engineers were among the first automated, because ML can automate ML.
- 3
Find the new categories it opens
Look for the new kinds of work the automation enables, the way telecom automation enabled podcasting.
- 4
Invest in choosing what's next
Put your human effort into defining what to do next — the part that remains a matter of human choice.
Pro tip This is exactly where computational thinking pays off.
In the wild
Wolfram recalls fears that automating telephone switching would erase all those operator jobs; the jobs did go away, but the automation opened many new categories — including the telecommunications infrastructure that makes the podcast they're recording possible.
→ Lost routine jobs were replaced by entirely new categories of human work.
150 years ago most Americans did agriculture by hand; machinery automated it and the fear was nobody would have anything to do, but the automation enabled many new types of work.
→ Employment shifted rather than vanished as new possibilities opened.
Common mistakes
Betting your career on routine skills
Learning low-level C, C++, Python or Java as a stable 'good job' misjudges the cascade — routine programming is exactly the crank-turning work that gets automated away.
Is it for you?
Best for
Workers and founders planning careers and strategy around advancing automation.
Not ideal for
Anyone seeking a guarantee that a specific current skill will stay economically relevant.
From the transcript
“once you've defined an objective you can build automation that does that objective... But then you have the question, well, what are you going to…”
“yes, those jobs went away, but that automation opened up many other categories of jobs.”
From the episode
Stephen Wolfram: How AI Works and How to Use It to Stay Ahead
Stephen Wolfram