From winning awards for technical advances in economic theory and industrial organization, serving as chief economist at Microsoft, conducting original research combining machine learning and econometric modeling, pioneering the new field of technological economy, having helped found and foster Stanford IHAand the launch of the Golub Capital Social Impact Lab at Stanford, Susan Athe will now try on a new hat as chief economist at the US Department of Justice’s (DOJ) Antitrust Division.
The varied nature of his work and career has earned him respect in many academic fields. “Susan is a force of nature. She moves from machine learning to business strategy, technology policy and social impact, producing deep insights at every turn,” says Jonathan Levinthe Professor Philip H. Knight and Dean of the Stanford Graduate School of Business, where Athey is an endowed professor in economics of technology.
Although Athey will continue her appointment to the GSB on a part-time basis as she enters her new role in government, the shift in focus offers an opportunity to reflect on the significant impact she has had throughout her tenure. career and at Stanford.
A model for HAI
Few researchers better exemplify the multidisciplinary mindset fostered by Stanford IHA than Athey. Her propensity to immerse herself in various fields of study dates back to her undergraduate years at Duke University, where she graduated in 1991 with a triple major in economics, mathematics, and computer science.
She then became a professor of economics and business at MIT, Harvard University, then Stanford from 2013, but even in the field of economics, Athey’s interests were diverse: in 2007 , she won the prestigious John Bates Clark Medal for his contributions to several subfields, including industrial organization, microeconomic theory, and econometrics.
But it was while on university leave to serve as Microsoft’s chief economist from 2008 to 2013 that Athey made a surprising connection between his passion for economics and the tools of AI and analytics. machine learning.
They already knew that digitization and technology platforms were going to play a big role in the economy and that search engines were about to take on outsized importance. She also knew that the research community was just beginning to grapple with questions about the design of digital markets and what healthy competition looked like in those markets, and she was excited to help develop this research.
But once she started working at Microsoft, Athey also discovered something she hadn’t expected: the potential of machine learning to solve economic problems. The creators of the Bing search engine were conducting experiments in ways economists could only dream of. They were simultaneously running thousands of random A/B tests – asking a lot of “what if” questions to better understand things like which search results should go to the top and how to run auctions to fix the prices of advertising on a search page. By comparison, she says, economists typically conduct one experiment a year.
“Microsoft was using an artificial intelligence system made up of hundreds of algorithms that all worked together to create a search results page,” she says. “It was something new.”
In economics until then, data mining and machine learning were pejorative terms for a less advanced form of statistics. “They were seen as a mechanical exercise to separate cats from dogs,” she says. But at Microsoft, Athey saw an opportunity to combine computational advances in predictive machine learning with statistical theory so that researchers could better understand causal effects not only in commercial applications like the search engine, but also in social sciences and economics. It was an epiphany that launched her in a new direction of research and helped define her as one of the first technology economists.
Machine learning and causal effects
Drawing on her experience at Microsoft, Athey realized that insights from predictive algorithms could be exploited in new ways by combining them with recent developments in econometrics and statistics. For example, machine learning algorithms could be adapted to answer cause and effect questions in economics, such as what will happen if we change the minimum wage? Developing immigration policy? Raise prices? Allow two companies to merge? “Predictive machine learning can’t solve these questions on its own, but it can help,” she says.
For example, Athey used machine learning to examine the impact on consumers of personalized pricing, a form of price discrimination that charges consumers different prices based on their willingness to pay. Traditional economic methods would provide global solutions to this problem, she says. Perhaps they would study one product category at a time, taking into account the demand, for example, for different brands of yogurt or napkins. By applying machine learning methods to consumers’ historical purchase data, Athey’s research group can estimate consumers’ personalized preferences across multiple products at the same time.
Building these predictive models of consumer choice, in turn, allows researchers to ask even bigger questions about things like what happens to prices if you apply a tariff or if generics come on the market. “To answer these questions, we want to understand how consumers make their choices,” says Athey. And machine learning provides that input in a way that allows researchers to do that work more efficiently, at scale, and in a more personalized way. “If you assume everyone is the same, it gives different answers than if you assume people have different preferences,” she says.
A pioneer of the technological economy
Athey’s role as chief economist at Microsoft ended in 2013, but her tenure there defined her as one of the first people to be considered a “technology economist.” It’s a field she has since helped establish as an independent discipline by convening the first conferences in the field and mentoring many students along this career path.
“Now the tech economists hold an annual conference that draws about 800 attendees,” she says. “And we have a specialized job market because being a technology economist is a distinct profession that people can pursue.”
Athe also has written on what it means to be a technology economist. “Part of it is a career, but it’s also a combination of different fields of study,” she says. Technology economists study the impact of digitalization on the economy, which involves thinking about market design, privacy, data security, fairness, competition policy, etc. ., she says. “They also help create and analyze business models and competitive strategy, and they connect models to data to guide decisions.”
Advancing AI for good
At Microsoft, in addition to unexpectedly delving into machine learning and AI, Athey witnessed first-hand the challenges created by these technologies – ethical and legal issues, First Amendment issues, fairness and bias. , privacy and copyright, and the prevalence of unintended consequences. as people manipulated or manipulated the system in response to market changes or new rules.
Because of these observations, Athey developed a desire to impact how machine learning and AI would play out in the world. When she returned to academia full-time, her first steps in that direction were helping plan the launch of HAI and then becoming one of HAI’s founding Associate Directors. “Stanford HAI was really created to solve these problems,” she says. “We want to make AI beneficial to humans, and we want to avoid all of these unintended consequences.”
Athey also wanted to translate successful uses of AI from the for-profit sector into the social impact sector. This desire led her to launch the Golub Capital Social Impact Lab at Stanford. “We bring the tech toolkit to social impact apps,” she says. So, for example, the Social Impact Lab conducted case studies on digital technology in education to improve student learning; developed and implemented approaches to target educational messages to maximize farmer engagement; developed and evaluated applications for digital tablets that guide nurses in advising patients; and developed methods to prioritize candidates for clinical trials of COVID-19 drugs.
Connecting the Dots at the DOJ
Applying machine learning to interesting social issues at the Golub Capital Social Impact Lab is a bottom-up approach to bringing about change, says Athey. By contrast, in her new position as chief economist of the DOJ’s antitrust division, she will try her hand at tackling the issues of the digital economy from the top down. “Laws and government policies affect everything from how competition works, to mergers, to the investments people make,” says Athey.
By moving to the DOJ, Athey hopes to continue many of HAI’s efforts to help governments adapt to a rapidly changing era of technology, particularly around the use of data in industry and government. “Because technologies such as artificial intelligence evolve so rapidly, it is difficult for the government to keep pace,” she says. “We need to understand how all branches of government are going to be ready to lead us through a different age.”
This is the perfect time for Athey to try his hand at government work, Levin says. “At a time when technology is booming and promoting competition is essential, I can’t think of anyone I’d rather have in the DOJ than Susan.”
Stanford HAI’s mission is to advance AI research, education, policy, and practice to improve the human condition. Learn more.