The 1st literature stream follows the logic of jobs automation, e.g. [Frey and Osborne, 2017] predict that automation may replace 47% of today’s jobs. It shows that AI improves the corporate governance and lowers the agency cost [Fenwik and Vermulen, 2018] by automating decision making using real-time Big-Data analysis [see e,g, Moll and Yigitbasioglu, 2019]. [Manita et al., 2020] show that AI benefits all governance parties: shareholders, BoD, auditors, etc. [Erel et al., 2018] show that a machine-learning outperforms a human in selecting the BoD members. [Issa et al., 2016] show that AI makes external audit more accurate and enables audit firms to focus on value-adding jobs. [Chan and Vasarhelyi, 2011] argue that machine processing creates opportunity for continuous auditing enabling BoD and shareholders to access the data in real time, not waiting for the regular reports. [Lombardi et al, 2015; Krahel and Titera, 2015] argue that big data improves the quality of financial statements while [Cunningham and Stein, 2018] argue that itanalysis helps with anomalies detection. [Wang et al., 2020] argue that machine learning helps identify risk factors and prevent corporate misbehavior. [Bae, 2012] argues that a more accurate prediction of financial distress can assist with the better decision making of CFO and boardroom and benefit investors. An adjacent literature stream covers “algorithmic governance” exploring full decision-making automation [see e.g. Danaher et al, 2017]. However, to the best of our knowledge, this stream is yet to cover the corporate governance. Despite the positive attitude there are of course AI skeptics. E.g. [Dignam, 2020] argues that AI may aggravate problems as discrimination, creates problems of liability attribution, etc. and should be treated with caution.

The researchers exploring the topic of organizational change argue that digital technologies transform the nature of a firm, making it less based on the traditional authority [Fenwick, McCahery, Vermeulen, 2019]. [Kiron and Unruh, 2018] argue that firms have to be much more inclusive of interests of customers. [Parker and Van Alstyne, 2016] highlight the importance of platform-based business models such as Uber, etc. [Fenwik and Vermulen, 2018] highlight that in digital technologies change “who, what, when, and how people “trust”. All the researchers agree that this environment calls for fundamental reconsideration of corporate governance, making it much more decentralized to reflect the changing nature of the business and for regulatory models’ revision. [Luna et al., 2014] argue for importance of Agile governance, while [Ansell and Gash, 2008] advocate collaborative governance.

Despite the consensus amongst researchers and implementations as making AI a part of BoD [Mosco, 2020], there are still multiple open questions: do AI exploring firms have better corporate governance and weaker principal-agent conflict? Do shareholders appreciate it, i.e., does investment in AI make the shareholders friendlier or more hostile towards a firm’s management? Do firms exploring alternative corporate governance benefit from it? What type of AI application is the best from the corporate governance point of view? What is the best way to proceed with AI implementation? These questions remain to be researched going forward.

This overview is a part of the research on implications of AI applications for corporate governance. The empirical results are coming shortly. The abstract contains references only to the selected sources used to demonstrate the gaps in the existing research. References list contains the full body of literature surveyed.