Knowledge that Transforms

Racial discrimination in crowdfunding is a significant barrier to equitable access to capital, as racial minorities face greater challenges in achieving their fundraising goals. While prior research has documented discriminatory patterns in crowdfunding outcomes, the underlying mechanisms driving this discrimination remain unclear. This is a critical gap that must be addressed to develop effective interventions. Drawing on economic theories of taste-based and statistical discrimination, we examine how discrimination mechanisms vary across crowdfunding types. We posit that donation crowdfunding primarily exhibits taste-based discrimination, while investment crowdfunding manifests statistical discrimination. We tested these predictions through three preregistered randomized experiments. Confirming our prediction, the first experiment revealed taste-based discrimination against Black fundraisers in donation crowdfunding. The second experiment demonstrated statistical discrimination in investment crowdfunding. However, contrary to prior research showing negative discrimination against minorities, our second experiment revealed positive statistical discrimination toward Asian fundraisers. We posit that this positive discrimination is due to the decision frame participants adopt, and this explanation is supported by Experiment 3: The direction of discrimination is reversed when task instructions are modified to elicit a negative decision frame. Our research advances both the crowdfunding and the racial discrimination literatures while providing insights for platform design and debiasing interventions.

Participants on tokenized platforms (i.e., platforms with blockchain implementation) can simultaneously take multiple roles, such as user, investor, and laborer, and draw income from the last two roles. Unlike traditional markets that typically prioritize one means of profitable participation, participants on such platforms need to allocate their efforts on the platform to increase revenue. We developed a decision framework for determining participants’ strategic participation on tokenized platforms to maximize earnings from investment and labor. Individual participants were distinguished from the platform-average participant, and decision-making is cast into two subproblems: (1) ignoring individual actions’ impact on platform state, we constructed strategies based on metrics that characterized model projections of future platform development and derived the metrics from Monte Carlo ensembles; (2) considering individuals’ actions as explicitly influencing the platform state, we formulated the control problem as a Markov decision process and solved it via reinforcement learning (RL). The framework addresses parameter uncertainty from model estimation, system uncertainty in model projection, and input uncertainty during participant-platform interaction. We compared metric-based and RL strategies from the two solution approaches using historical token price series; the results suggest good performance of our decision framework.

How does online trading intensity influence investors’ gambling preferences? How do financial advisors shape the relationship between online trading intensity and gambling preference? This study answers these important questions in response to concerns about whether digital technologies such as the Internet exacerbate the tendency of retail investors to gamble in the stock market. Leveraging transactional trading data of approximately 20,000 investors and more than 450,000 monthly observations, we find a U-shaped relationship between online trading intensity and investors’ holdings of lottery-like securities. Interestingly, we show that financial advisors flatten this relationship between online trading intensity and investors’ holdings of lottery-like securities. In other words, financial advisors can reduce the undesirable costs associated with online trading. These findings, which draw on literature at the intersection of information systems and behavioral finance, provide valuable insights on how the Internet influences stock trading. Our work suggests the continued relevance and importance of human capital (e.g., financial advisors) in the digital era increasingly shaped by artificial intelligence, and provides guidance as to when financial advisors are more effective in reducing investors’ gambling preferences.

In the sharing economy, screening can mitigate information asymmetry between service providers and consumers, but excessive screening may cause market inefficiency. We adopt a multimethod approach to examine how forgoing screening affects competition in the sharing economy. We first develop a game-theoretic model to analyze how a focal service provider’s decision to forgo screening impacts its own and a rival provider’s prices, market demands, and profits. Using Airbnb as the investigation context, we then empirically validate our analytical results. We find that our key analytical and empirical findings are consistent. Specifically, due to aggressive responses from rival providers, a low-rating focal provider may lower its service price after forgoing screening despite its increased attractiveness to consumers. Intriguingly, forgoing screening cannot increase a provider’s demand if its rating is high, whereas the demand for rival providers will increase. Moreover, our analysis reveals that in markets with relatively high competition intensity, forgoing screening benefits all three parties—the focal provider, rival providers, and the sharing platform—thereby improving the total social welfare. Based on these findings, we suggest that sharing platforms should encourage high-rating providers to forgo screening as a strategic approach to enhance the overall market performance. Furthermore, special attention is needed to prevent a decline in social welfare when the provider forgoing screening has a low rating, while its rivals have high ratings. Our findings reveal that platform governance through feature-level design can directly enhance platform profits at the expense of social welfare, thereby creating misaligned incentives and calling for external regulation.

Previous research on recommender systems primarily focuses on their short-term effects on customer search and purchase behaviors. This study investigates the effect of a recommender system on customer loyalty and long-run retail sales using a randomized field experiment. We manipulated the presence vs. absence of product recommendations from an item-based collaborative filtering recommender system at an online retailer. The results reveal that displaying recommendations increases consumers’ purchases of recommended products but at the cost of reduced sales of non-recommended ones, which may not increase total sales in the short run. In spite of this, consumers’ shift of focus (to recommended products) due to the presence of recommendations plays an important role in enhancing customer loyalty. When recommendations are disabled, long-term sales decrease significantly due to reduced customer loyalty. The results show that the loyalty effect is primarily driven by a preference effect, in which displaying recommendations induces consumers to view more recommended products, enhancing their shopping experience and increasing customer returns. For returning visitors, who may deem product recommendations as built-in elements of a desirable store environment, a one-time disablement of recommendations can directly lead to defection. The findings provide insights on the true value of recommender systems.

Doxing, the malicious exposure of personal information online to threaten or intimidate individuals, is an escalating global threat. A particularly consequential form is politically motivated doxing (PMD), where perpetrators— known as doxers—target individuals for their viewpoints and beliefs, shaping sociopolitical discourse in the process. Despite PMD’s profound impact, research on the perspective of victims (referred to as doxees) remains scarce. Drawing on affective events theory, this study argues that doxees perceive PMD as a breach of their fundamental civil rights. We theorize that PMD triggers an affective duality—fear and anger—in doxees, which drives divergent behavioral shifts in their civic engagement across online and offline domains. Additionally, we examine how doxees’ strength of conviction regarding the PMD issue moderates their affective reaction, revealing distinct responses among individuals with moderate versus extreme views. Using a monostrand mixed-methods approach with a dominant quantitative and a complementary qualitative component, we conduct a personalized, immersive online vignette study in the context of U.S. school boards to empirically test our model. Our findings contribute to the IS literature by: (1) advancing theoretical and empirical understanding of doxees’ perceptions of PMD, (2) linking these perceptions to their affective and behavioral responses, and (3) demonstrating that PMD has a particularly suppressive effect on civic engagement among doxees with less firmly held convictions, highlighting PMD as a mechanism through which malicious actors exacerbate sociopolitical polarization. Our work offers timely implications for policymakers and online platforms.

This paper examines the influence of artificial intelligence (AI) on research and development (R&D) practices, proposing that AI enables firms to discover, evaluate, and make sense of information from their partners, thereby reducing information asymmetry and allowing alliances to flourish. Empirically, we approximate a firm’s AI resources using patents and job postings to analyze the relationship amongst AI, alliances, and drug innovation in the pharmaceutical industry. Our findings show that firms with greater AI resources shift their locus of innovation toward alliances and develop more drugs from their alliances. Furthermore, AI can be particularly useful in exploiting information held by counterparties within an alliance. Taken together, our study highlights the complementary role of AI and alliances, shedding light on how modern organizational and technological advancements jointly shape the production of innovation.

This paper explores how different types of fact-checkers (i.e., AI or human) impact users’ perception of the believability of news that is flagged as false. Building on source credibility theory, we evaluate how the reputation of the news source and the user's political orientation (i.e., progressive or conservative) moderate the impact of AI versus human fact-checkers. We examined this interaction in two separate 3×2×2 online quasi-experiments conducted in the United States and the United Kingdom. In both studies, we found differences in the impact of fact-checker type and in the moderating effects of poster reputation and user political orientation. Our results show that AI fact-checkers are more effective than human fact-checkers in reducing users’ perceptions of news believability, particularly among progressives. We also found that if the news poster has a high reputation, this can further enhance this impact. By investigating the interplay among fact-checker type, the poster, and users’ political orientation, and by comparing results across two countries, we extend understanding of how different fact-checker types affect news believability regarding false news on social media platforms (and user engagement in robustness checks). Finally, we derive managerial implications for mitigating the spread of false news on social media platforms.

This study examines how the extent of business intelligence (BI) systems within business enterprises is associated with compliance risk. BI tools promote a risk-aware culture by providing real-time data and analytics, helping establish objectives aligned with compliance, and ensuring transparent communication of compliance information across the organization. Using BI, organizations can better address compliance challenges, strengthen risk management practices, and adhere to regulatory requirements more effectively. Using a BI implementation dataset, we find that firms with more widespread BI implementation in the organization have lower compliance risk, reporting fewer regulatory violations and lawsuits. We also provide evidence that some measures of strategic IT focus moderate the association between BI implementation and compliance risk outcomes. Delving into the particularities of the association between BI implementation and compliance risk, we find that greater BI implementation is associated with lower lawsuit settlement amounts, fewer lawsuit types, and shorter lawsuits. Further, the beneficial BI implication is associated mainly with operational compliance risk rather than financial accounting compliance risk. Our findings have crucial implications for the value of BI in corporate compliance risk management.