Kannan Srinivasan

Behnam Mohammadi, Nikhil Malik, Tim Derdenger et al.

Recent AI algorithms are black box models whose decisions are difficult to interpret. eXplainable AI (XAI) is a class of methods that seek to address lack of AI interpretability and trust by explaining to customers their AI decisions. The common wisdom is that regulating AI by mandating fully transparent XAI leads to greater social welfare. Our paper challenges this notion through a game theoretic model of a policy-maker who maximizes social welfare, firms in a duopoly competition that maximize profits, and heterogenous consumers. The results show that XAI regulation may be redundant. In fact, mandating fully transparent XAI may make firms and consumers worse off. This reveals a tradeoff between maximizing welfare and receiving explainable AI outputs. We extend the existing literature on method and substantive fronts, and we introduce and study the notion of XAI fairness, which may be impossible to guarantee even under mandatory XAI. Finally, the regulatory and managerial implications of our results for policy-makers and businesses are discussed, respectively.

Mi Zhou, Vibhanshu Abhishek, Timothy Derdenger et al.

This study analyzed images generated by three popular generative artificial intelligence (AI) tools - Midjourney, Stable Diffusion, and DALLE 2 - representing various occupations to investigate potential bias in AI generators. Our analysis revealed two overarching areas of concern in these AI generators, including (1) systematic gender and racial biases, and (2) subtle biases in facial expressions and appearances. Firstly, we found that all three AI generators exhibited bias against women and African Americans. Moreover, we found that the evident gender and racial biases uncovered in our analysis were even more pronounced than the status quo when compared to labor force statistics or Google images, intensifying the harmful biases we are actively striving to rectify in our society. Secondly, our study uncovered more nuanced prejudices in the portrayal of emotions and appearances. For example, women were depicted as younger with more smiles and happiness, while men were depicted as older with more neutral expressions and anger, posing a risk that generative AI models may unintentionally depict women as more submissive and less competent than men. Such nuanced biases, by their less overt nature, might be more problematic as they can permeate perceptions unconsciously and may be more difficult to rectify. Although the extent of bias varied depending on the model, the direction of bias remained consistent in both commercial and open-source AI generators. As these tools become commonplace, our study highlights the urgency to identify and mitigate various biases in generative AI, reinforcing the commitment to ensuring that AI technologies benefit all of humanity for a more inclusive future.

Yue Qiao, Kannan Srinivasan, Anish Arora

Existing secret key extraction techniques use quantization to map wireless channel amplitudes to secret bits. This pa- per shows that such techniques are highly prone to environ- ment and local noise effects: They have very high mismatch rates between the two nodes that measure the channel be- tween them. This paper advocates using the shape of the channel instead of the size (or amplitude) of the channel. It shows that this new paradigm shift is significantly ro- bust against environmental and local noises. We refer to this shape-based technique as Puzzle. Implementation in a software-defined radio (SDR) platform demonstrates that Puzzle has a 63% reduction in bit mismatch rate than the state-of-art frequency domain approach (CSI-2bit). Exper- iments also show that unlike the state-of-the-art received signal strength (RSS)-based methods like ASBG, Puzzle is robust against an attack in which an eavesdropper can pre- dict the secret bits using planned movements.