New research from Harvard Business School highlights some of AI’s biggest assumptions

AI For Business


Harvard Business School faculty this week released a series of new research papers and case studies that squarely confront some of the biggest debates in business today: the hype and reality of AI, the growing influence of AI-generated financial news, geopolitical fragmentation, and the limits of globalization.

Among the most talked about papers is a revised HBS working paper that takes a skeptical but pragmatic perspective on the use of large-scale language models in market research. In “Using LLM for Market Research,” Professor James Brand, Professor Ayelet Israel, and Professor Donald Ngwe conclude that while AI tools may approximate human survey responses, they can also produce inaccurate or “false signature” results when predicting consumer preferences.

The authors argue that LLM is best used as a supplement to human research, rather than a replacement for it. They found that the strongest results were obtained when the model was fine-tuned using prior human survey data from similar product categories and customer groups.

AI may change the way investors read news

Another new working paper from HBS explores how AI-generated summaries are already reshaping investor behavior.

In “Financial Media Generation AI and Investor Processing,” Professor Tony Cho, Professor Allen Huang, Professor Joseph Parsley, and Professor Christina Rennekamp analyze the evolution of AI summaries. wall street journal And we found that articles with AI-generated summaries elicited higher trading volumes and stronger market reactions.

The researchers also take issue with a common concern about AI overviews: that they encourage shallow or skimming reading. Rather, their experiments suggest that summaries may actually improve comprehension and recall of full-text articles and direct investors’ attention to the highlighted information.

The study concludes that AI-generated summaries are becoming “interpretive cues” that shape how investors consume and react to financial news.

Greece’s National AI Strategy

HBS also released a new case study examining how governments seek to leverage AI as part of their national economic strategy.

“Greece’s Great Leap Forward? AI as a Catalyst for Education, Entrepreneurship, and National Prosperity” by George Serafeim, Deborah Spahr, and Nicole Zelasko focuses on the partnership between Greece and OpenAI to bring ChatGPT Edu to Greek secondary schools and launch a startup accelerator supported by OpenAI technology and mentorship.

The case frames AI not just as a productivity tool, but as a potential means to reverse brain drain, modernize education, and accelerate entrepreneurship in countries still trying to rebuild after years of economic instability.

Questions about American financial control

Another timely paper is by Professors Wenxing Du, Lit Keelati, and Jesse Schrager, who argue that the long-assumed relationship between the dominance of the U.S. Treasury and the dominance of the dollar may be weakening.

A forthcoming IMF Economic Review paper, “Decoupling Government Debt and Dollar Privileges,” argues that while the dollar has maintained its status as a global “convenience” and safe-haven asset since the global financial crisis, the relative attractiveness of U.S. debt has declined and even turned negative in some maturities compared to other developed country bonds.

The paper comes as economists and policymakers increasingly debate the long-term sustainability of U.S. financial leadership amid rising debt levels and geopolitical divisions.

Why retailers fail overseas

Meanwhile, another HBS research report argues that many retailers misunderstand why international expansion efforts fail.

In “Retail Expansion into International Markets: Why Some Retailers Succeed and Many Fail,” Professors Srikant Gokhale and Rajiv Lal argue that the biggest problem is not localization errors after market entry, but rather a failure to identify in advance which parts of a company’s business model are essential and which can be adapted.

“The strategic mistake is not insufficient localization,” the authors write. “There is a lack of balanced adaptation.”

Don’t miss Harvard University launches ‘AI Foundry’ to make the case for going beyond case studies

The post New research from Harvard Business School takes aim at some of AI’s biggest assumptions first appeared on Poets&Quants.



Source link