Edward Conard

Top Ten New York Times Bestselling Author

  • “…serious thinking for serious thinkers. …a thought-provoking blueprint for growing middle- and working-class incomes.” - Mitt Romney, former Governor of Massachusetts
  • “Unintended Consequences is full of substance, it is one of the must-read books of the year, and once I finish it I will be giving it a second read through right away.” - Tyler Cowen, Professor, George Mason University
  • “Unintended Consequences provides a provocative interpretation of the causes of the global financial crisis and the policies needed to return to rapid growth. Whether you agree or not, this analysis is well worth reading.” - Nouriel Roubini, New York University; Chairman, Roubini Global Economics
  • “Unintended Consequences is far smarter and more thought-provoking than most economics written for the general public” - Greg Mankiw, Harvard University, Former Chairman of the Council of Economic Advisors
  • “A full-throated defense of economic dynamism.” - The Wall Street Journal
  • “Unintended Consequences represents the most cogent and persuasive analysis of the Financial Crisis to date.” - Andrei Shleifer, 1999 John Bates Clark Medal Winner
  • “…reminds us that inequality sends a signal of what society lacks most, in America’s case, entrepreneurship and risk taking.” - Lawrence Lindsey, CEO, The Lindsey Group, former Director of the National Economic Council
  • “…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
  • “…reminds us that inequality sends a signal of what society lacks most, in America’s case, entrepreneurship and risk taking.” - Lawrence Lindsey, CEO, The Lindsey Group, former Director of the National Economic Council
  • “There are an amazing number of good ideas and interesting points made in Unintended Consequences. The thinking underlying it, and the obvious depth of understanding of the author, are very impressive.” - Steven Levitt, coauthor of Freakonomics; 2004 John Bates Clark Medal
  • “Unintended Consequences should be read by anyone who takes for granted the superiority of progressive taxation and has not thought carefully about the trade-offs involved.” - The New Republic
  • “…a must-read for serious students of economic policy.” - Glenn Hubbard, Dean, Columbia Business School, and former Chairman of the Council of Economic Advisers
Upside of Inequality Oxford Unintended Consequences

AI Summary. In an experiment with over 500 professionals across management, data analysis, and consulting using 13 different large language models, each year of model progress cut task time by 8%, and a 10x increase in training compute is associated with a 6.3% time reduction. The authors project continued scaling could raise US productivity by about 20% over a decade.

Ali Merali Yale University
Date Posted:
Is Database:
Database

An experiment with over 500 professionals found that each year of model progress cut task time ~8%, with 56% of gains driven by increased compute and 44% by algorithmic progress, suggesting that US productivity could rise 20% over the next decade.

How much does AI model scaling raise professional productivity?

Core argument: In experiments with over 500 professionals across management, data analysis, and consulting using one of 13 LLMs, each year of model progress reduced task time by 8%, with 56% of gains from added compute and 44% from algorithmic progress.

This paper derives “Scaling Laws for Economic Impacts”—empirical relationships between the training compute of Large Language Models (LLMs) and professional productivity— [from] over 500 professionals across three high-skill domains: management, data analysis, and consulting [using] one of thirteen different LLMs. Each year of model progress reduced task time by 8%, with 56% of gains driven by increased compute and 44% by algorithmic progress; a tenfold (10x) increase in model training compute is associated with a 6.3% reduction in time taken. Access to any AI model increased base Earnings Per Minute (EPM) by 81.3% (p = 0.001) [and] resulted in a 146% increase in Total Earnings Per Minute (TEPM), with AI assistance delivering total earnings gains of $1.58 per minute on non-agentic, analytical tasks [versus] $0.34 per minute for “agentic” tasks requiring multi-step interactions with external tools. While the quality of autonomous model output scales linearly with compute, [for human-assisted work] the scaling law for quality completely disappears. Multiplying the share of tasks that can be automated (19.9%), multiplied by the average productivity effect derived from our scaling projections (175.1%), multiplied by the labor share of costs (57%) yields Total GDP Gain ≈ 19.9%, [meaning] continued model scaling could boost U.S. productivity by ~20% over the next decade.

Takeaways by Macro Roundup® AI

  1. In experiments with over 500 professionals across management, data analysis, and consulting using one of 13 LLMs, each year of model progress reduced task time by 8%, with 56% of gains from added compute and 44% from algorithmic progress.
  2. Access to any AI model raised base Earnings Per Minute by 81.3% (p = 0.001) and Total Earnings Per Minute by 146%, with gains of $1.58 per minute on non-agentic analytical tasks versus $0.34 per minute on agentic tasks requiring multi-step tool use.
  3. The authors multiply the automatable task share (19.9%) by the projected average productivity effect (175.1%) and the 57% labor share of costs to project that continued model scaling could boost U.S. productivity by roughly 20% over the next decade.

Related Articles:

  • US Productivity Accelerates to Fastest Pace in Two Years — US nonfarm productivity, or employee output per hour, rose at a 4.9% annualized rate following an upwardly revised 4.1% gain the prior quarter, according to the Bureau of Labor Statistics. Unit labor costs fell 1.9% after a prior-quarter decline, the first back-to-back drop since 2019.
  • Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence — Using payroll records covering millions of US workers, the authors find employment for workers aged 22–25 in occupations most exposed to artificial intelligence, such as software developers and customer service representatives, declined 6%, while employment for older workers in those occupations rose 6–9%. Employment in less-exposed occupations grew comparably across age groups.
  • Weekend Thinking: A Cul-de-Sac With a View — Leading AI developers are shipping smaller, more frequent model updates, creating an impression of momentum through release frequency rather than the magnitude of change. The author argues the bet that scaling data, compute, and model size would produce artificial general intelligence has not paid off, pushing labs back toward harder problems like continual learning and new architectures.
Showing 78 database articles primarily about Innovation/Research
Currently filtering by: