AI can extend expertise to a larger set of workers
In this paper, David Autor explains how the fear that AI will destroy jobs is misplaced. On the contrary, he argues, AI has the potential to restore the middle-class jobs in the U.S. that have been hollowed out by automation and globalization. This is because AI if used well, can assist in extending the relevance, reach, and value of human expertise for a larger set of workers by enabling them to perform higher-stakes decision-making tasks currently reserved for elite experts. Autor discusses the historical transformation of expertise from artisanal to mass expertise and then to elite expertise in the Information Age, highlighting the potential of AI to reshape the value and nature of human expertise.
The unique opportunity that AI offers humanity is to push back against the process started by computerization — to extend the relevance, reach, and value of human expertise for a larger set of workers.
Applying AI to Rebuild Middle Class Jobs
Authors: David H. Autor
From: MIT
How to use large Language Models in Economics
Anton Korinek in this paper identifies six different areas where LLMs can be useful in economics:
- Ideation: LLMs can assist in brainstorming, evaluating ideas, and providing counterarguments.
- Writing: They can synthesize text, provide examples, edit and evaluate text, and generate titles or tweets.
- Background Research: LLMs can be used for searching and summarizing the literature, translating text, explaining concepts, and formatting references.
- Coding: They are capable of writing code based on natural language instructions, explaining code, translating code between programming languages, and even debugging code.
- Data Analysis: LLMs can extract and reformat data from text, classify text, extract sentiment, and even simulate humans to generate data.
- Mathematical Derivations: They can assist in setting up models, working through derivations, and explaining models.
Language Models and Cognitive Automation for Economic Research
Authors: Anton Korinek
From: University of Virginia
