Notes on Average is Over: Powering America Beyond the Age of the Great Stagnation by Tyler Cowen
Part 1: welcome to the hyper meritocracy
1 work and wages in iWorld
Diminished job opportunities
Imbalance in tech growth between industries
Key question: are you good at working with intelligent machines
Are u competing against a computer or is it helping u compete
Intelligent machine analysis getting good
Voice analysis to help with negotiations
2 big earners and losers
What is scarce: quality land and natural resources, ip, quality labor with unique skills
Not scarce: unskilled labor, capital (money in bank earning no return)
Drones create jobs; requires 168 ppl to keep drone running for 24 hrs
Labor that can augment new tech will earn well
Marketing as seminal sector in new economy
Making ppl feel better helps sell more
Competition for attention of high earners
Top earners from Internet and hedge fund sectors
Knowing how to spot and recruit those who are good at working with mechanized intelligence
The better the world gets at measuring value, the more competitive it will become (hyper meritocracy)
Acquihiring
Engineers worth 500K-1M
Managers play important role in coordinating complex processes
One bad employee can ruin big complex system
Women more conscientious and reliable so will get more critical and personal service jobs where importance on reliability
Monomaniacal commitment and extreme level of skill will still go more to men
Important skills: exactness of execution over accumulation of brute force, consistent coordination over time, morale to motivate production and coordination
Capital surrounding a worker is up
Easier to destroy than create
Picky, snobbish, elitist
Consulting, law, and finance are best careers
Consulting because long supply chains complex and good for consultants to optimize
Law because lots of new regulation and litigation
Finance because promise of bailouts promotes risk
Smart but underspecialized ppl need to learn things so these are good places to do it
Young contributed to Internet industry without training because was immature industry
Sometimes the general intelligence used for consulting does provide value to overspecialized businesses
3 why are so many ppl out of work
At first machine hardly adds anything beyond what human does
Then machine gets ok but needs help of expert humans to work optimally
Then machine works well and only needs low skilled humans for help
Finally machine works best by itself and surpasses humans and humans move on to help machines in industries less far along
Machines as complements to humans and not replacements
Almost symbiotic relationship
Not about having computer mimic human brain better
Computers can excel at chess because of regularized environment but still lag human in things like poker which involves psychology, reading ppl, bluffing
Effective man machine teams of future
Fewer Americans trying to work
Male real wages falling over long run
Great Recession
A lot of workers were overplayed relative to their skills
Many zero marginal product workers who u would never hire even for 0 wage
3/4 of youth not accepted to serve in military even
Per labor hour productivity
Fewer jobs left in the middle and lower wage ones mostly overseas
Regulations and benefits increasing the cost of hiring so don't want to do it for low wage workers
Higher min wage and more employee litigation hurting hiring
Freelancers in generation limbo
Many starting own businesses because can't get jobs
Micro jobs
Personal Service jobs
Part 2 what games are teaching us
Ch 4 new work old game
Gaming industry bigger than Hollywood
Learning from how humans play chess and interact with machines
Window into how high earners will work with machines
Machine games
Watching 2 computers play against each other
Ugly counterintuitive computer moves beat grandmasters
Humans learning from computers on moves and strategy
Computers may guide social and financial decisions one day but will create stress when nonintuitive advice given and many won't like it
Gains will go to those who can manage the stress
Machines telling us how to date and do business in real time
Working with machines
Freestyle chess (human plus machine) is model for high earner of future
Ch 5 our freestyle future
Freestyle chess allows consulting computer, books, calling ppl, anything
But time limit is tight
Best players don't have high individual ratings but are good at aggregating advice from many chess engines
Human role very little during game but very big in preparation in between
Problem with grandmasters: not enough trust in the machines
Machines r us
Deep opening book
Human can prepare surprise innovations before performance or interview and then pull them out when useful and have advantage
skill: human memory of computer generated advice
Freestyle masters
Medical diagnosis based on human and AI working together
Doctors doesn't have to be as skilled
Needs to be good at knowing weaknesses of machines and how best to work with then
Most frequently used doctor in world is google
Ppl type in symptoms, check results, and then decide if should see doctor
Making freestyle work
Don't need more education
Need better skills in working with machines
Troubleshooting
Lessons
Human computer teams are the best teams
The person working the smart machine doesn't need to be an expert at the task at hand
Below some level of skill adding a man to the man machine process will make it less effective than the machine alone
Knowing one's limits is more important than it used to be
Zero marginal product worker is one who makes the outcome worse than if it was just the machine alone
A valuable worker has the potential of improving on the product of just the machine alone
The productive worker and the machine are stronger complements than before
Chapter 6 white your intuition isn't helping you get a job
Online dating services present matches that don't match intuition
eHarmony can nudge us from our indecisiveness and to question our own biases
Machines have no fear of unfamiliar
Decision making
Computer chess has contempt aversion: avoiding draws and seeking out unfamiliar complications
Unintuitive piece placements
Signs of improvement
Humans can learn and develop new talents
Intelligence level increasing
Future of intuition
Human strengths and weaknesses predictable
Be skeptical of the elegant and intuitive theory
It's harder to get out of your own head than u think
Revel in messiness
We can learn
7 the new office: regular stupid and frustrating
Telephone help menus getting a lot smarter and a lot stupider at the same time
Digital and physical environment
GPS
Driving direction errors due to real world street issues
Will world become more regularized and stupid so it's friendlier for machines?
Grocery self checkout
Machine assessments
Machines and workers assessing workers
Yelp reviews for all
Lawyer evaluations by machines
Duds will try to avoid ratings by machine
Doctors rating patients and filtering for most compliant
Customer reputation in stores
Our legitimate fears
8 why the Turing game doesn't matter
The long run
Better to create machines that complement human intelligence and not try to mimic it
Man Machine convergence
Turing test not interesting for man machine teams
Imitation not right standard
Autistics might fail Turing test
Many ppl cheating by using robots
Opposite was mechanical Turk
Identifying cheating based on similarity To machine not accurate
Human skills specializing in areas where machines aren't good
Memory and search
Google crutch
Lose mental skills
Science of memory gone
Was trying to make memories searchable
Doesn't mean we become stupider
Have better background info for decisions
Become better at searching for answers
Become more spontaneously creative
The machines place
Humans reticent to accept personal advice from machines or like the music they compose
Part 3 the new world of work
Ch 9 the new geography
Outsourcing and foreign competition not the culprit
Tech plays bigger role in compensation shifts than foreign trade
Immigration not culprit
Outsourcing and immigration
Hiring in USA getting more expensive due to regs so entrepreneurs outsourcing more
Lowers wages but brings cheaper goods
A global geographic trend
Well educated high earners will move to regions with best firms
Regions of europe on the periphery will be just for retirees and tourists and will cease being productive
The good of foreigners
10 relearning education
Competitive education through moocs, blogs, ted, apps
Learning games
Gaming tech already figured out how to educate best
New higher ed models
Hypermeritocracy: can learn from anywhere but only most talented and disciplined will rise ahead
Face to face teaching
Many tasks replaced by computer
Now specializes in teaching psychology behind skills, motivating, teaching how to best use computer, being role model
Makes chess instruction FaceTime skills harder to outsource because more about the person skills
Professors will just be motivators and coaches like for athletics and therapy
11 the end of average science
Increase in incomprehension
Problems becoming more complex
Contributions becoming more specialized
Machines will become own researchers
Specialization
Lone contributions harder to maintain
But machine intelligence also creating new opportunities
Bureaucracy and regulation killing science
Impossible problems
Breakthroughs by middle aged more than young but they are less willing to defy status quo
Machine science
Human feeding computer to do its research and then interpreting it
Less intelligible theories when machines create them
Whither economics
Data crunching instead of theoretical intuition
Power and qualify of data will grow faster than our models
Journal articles will become machine readable inputs for further machine research
Computer still bad at data gathering
12 a new social contract
Bigger gap between rich and poor
But cheap fun and education for masses
More fending for yourself
Self motivation predicting upward mobility
Meritocracy
The fiscal crunch
Zoning and new healthcare laws will drop wages and cause ppl to move out to cheaper states like to Texas
Ppl want more cash in their pocket so prefer to live cheaply
Build shantytowns to house old and poor by their choice
Living off the grid in own trailer
Budgets will get tighter
Medical rationing will increase
Ppl with good self control and motivation will live longer
Politics of the future
Older population and more conservative
More like Florida
Economically left behind will be more conservative