Notes on Automation
The topic of automation has been top of mind as of late, both in client work, and public discussion. I wanted to collate some material and highlight interesting views.
Automation Nation, CBS Sunday Morning
Great segment on CBS’ Sunday Morning Show, looking at the use of algorithms in trading, driving, retail and more.
According to one research up to 47% jobs could be lost to automation in the next 20 years,
Absolutely love this thought experiment by David Autor
It’s the year 1900 and 40% of all employment is in agriculture, so some economist from MIT teleports back in time to a farmer and tells him that a 100 years from now only 2% of people would be working in agriculture , what do you think the other 38% are going to do? Would he say “oh…search engine optimization”, “health and wellness”, “software”, “mobile devices. Most of what we do barely existed 100 years ago.
And the narrator adds:
Just because we can’t predict what we will be doing, it does not mean we will be doing nothing
Good anecdote from Elisha Weisel (Goldman Sachs)
” … and all of a sudden that young person is engaging with the client on their actual problems, rather than being stuck until 1am doing nothing but manning several different spreadsheets and trying to corral all of this data together”
Actual problem sounds to me like asking the right question, if all answers are known.
Further links mentioned in the piece:
EAC Panel on AI and Jobs
CBMM External Advisory Committee Member David Siegel (TwoSigma) talks with MIT principal research scientist Andrew McAfee about the role of artificial intelligence and jobs.
Lastly, here is my own (summarized) view on the topic of narrow AI, and its relevance to job security and automation.
- the brain is not computable
therefor artificial general intelligence is a fallacy
- if a task can be described by set of instructions
then it will be performed by a machine, and will exponentially improve to an unknown end
- Narrow AI cannot leave its domain of operation, only humans can
- therefor if what you do can be broken into sets of instructions you will be replaced by a machine