Ongoing discussions - articles, current events, etc

[donbowen]

I'm going to pin this discussion as a place everyone can discuss and share things that interest them and are (probably?) relevant to the course.

• Feel free to share any student events, conferences, or club news
• Share news! I'll post some news from the last year below to get the sharing going.
• Humans can learn from AI. Here is a cool study finding that since AlphaGo was released, human players have gotten significantly* better. More cool stuff in the paper disentangling the mechanism and how it’s changed play at the highest levels.

[donbowen]

Using CAPM to get a discount rate for free cash flow evaluations empirically seems to perform poorly!

A new working paper. Read it and then go bother my colleagues teaching you and your friends about FCF! (Or, use it and cite it to crush your competition at your internship.)

It starts like this:

The CAPM is the cozy bedtime story that tells students and practitioners that the world is
in good order and that they have learned something which will allow them to understand
it. But the real world isn’t like that. (Welch, 2021)

Lol.

---

Super interesting!

A study of FinTech lending reveals that loan rates still heavily depend on traditional credit scores, resulting in pricing distortions that lead to transfers from nonprime to prime and low- to high-risk borrowers, from Johnson, Ben-David, Lee, and Yao https://nber.org/papers/w31154

Here is the paper: https://nber.org/papers/w31154

Tweet: https://twitter.com/nberpubs/status/1650560395289976833

---

ABCD (always be checking your data) is the way, per the president and cofounder of OpenAI :

Manual inspection of data has probably the highest value-to-prestige ratio of any activity in machine learning.

Source: https://twitter.com/gdb/status/1622683988736479232?s=12&t=tQFtSoPd8xkIwSDT8QeONw

[donbowen]

Good programming habits, from https://twitter.com/almormd/status/1553303559818645504?s=51:

QA walks into a bar:
Orders a beer.
Orders zero beers.
Orders a million beers.
Orders batman and -1 beer.
Orders a peanut and a Mandelbrot fractal.

A real customer walks in and asks to use the toilet. Bar explodes, killing everyone.

@LeDataSciFi/classmates-2024

[donbowen]

hi @LeDataSciFi/classmates-2024

Here is a cool info graphic of ML, AI & Data Landscape. It's a ginormous graphic of firms, divided by the kind of work they do and the area they work in. I had to zoom out and scroll to see it all. It's a very dynamic space!

[donbowen]

The Labor Impact of Generative AI on Firm Values

How do recent advances in Generative AI affect firm value? We construct the first measure of firms' workforce exposures to Generative AI and show that an "Artificial-Minus-Human" (AMH) portfolio that is long high-exposure firms and short low-exposure firms earned daily returns of 0.44% in the two weeks following the release of ChatGPT. The labor-exposure effect is more pronounced for firms with greater data assets and is distinct from the effect of firms' product exposures to Generative AI. Highly-exposed workforces can be either substituted for or complemented by Generative AI technologies. We measure whether the exposed tasks are core or supplemental to assess relative substitutability. Examining firms' labor demand and profitability following the release of ChatGPT supports a labor-technology substitution channel for the increases in exposed firms' values that we document.

This is from a research paper here.

[donbowen]

[donbowen]

A really cool new article from the US Census and John Haltiwanger tracking AI use by firms: "Tracking Firm Use of AI in Real Time: A Snapshot from the Business Trends and Outlook Survey"

Abstract:

Timely and accurate measurement of AI use by firms is both challenging and crucial for understanding the impacts of AI on the U.S. economy. We provide new, real-time estimates of current and expected future use of AI for business purposes based on the Business Trends and Outlook Survey for September 2023 to February 2024. During this period, bi-weekly estimates of AI use rate rose from 3.7% to 5.4%, with an expected rate of about 6.6% by early Fall 2024. The fraction of workers at businesses that use AI is higher, especially for large businesses and in the Information sector. AI use is higher in large firms but the relationship between AI use and firm size is non-monotonic. In contrast, AI use is higher in young firms although, on an employment-weighted basis, is U-shaped in firm age. Common uses of AI include marketing automation, virtual agents, and data/text analytics. AI users often utilize AI to substitute for worker tasks and equipment/software, but few report reductions in employment due to AI use. Many firms undergo organizational changes to accommodate AI, particularly by training staff, developing new workflows, and purchasing cloud services/storage. AI users also exhibit better overall performance and higher incidence of employment expansion compared to other businesses. The most common reason for non-adoption is the inapplicability of AI to the business.

Link to full paper: https://www.nber.org/papers/w32319

@LeDataSciFi/classmates-2024

[donbowen]

https://x.com/culturecrave/status/1780508777248338016?s=46

Good stuff

[dannyhubert9]

https://www.youtube.com/watch?v=m6xCXSLscfU&list=WL&index=1&t=170s

Here's a really great podcast from Pablo Torre, a former ESPN correspondent, on the topic of AI with Josh Tyrangiel, a former deputy managing editor of TIME magazine as well as a former editor at Bloomberg Businessweek. Gives a really good overview of the key players in the AI space and where everything is headed. Highly recommend it if you want something to listen to while working on ASGN 8! Great extension of our conversation from Wednesday's class.

[donbowen]

I've enjoyed this so far, thanks!