By Alan Hewitt
In previous blog pieces I have written about the opportunities and threats posed by AI, both for companies and their boards. Clearly AI will open up great possibilities for boards to be better informed, leading, one hopes, to better decisions. I stressed that the use of AI needs to be assessed against What is Possible, What is Feasible and finally What is Desirable [see here].
Taking this to its logical conclusion I thought I’d look at whether it is possible for an AI enabled “robot” to be appointed to and become a fully contributing part of a company board.
It would seem that it didn’t require the current focus on AI to identify the first, almost real, example.
I found this article about a Hong Kong company called Deep Knowledge Ventures who in 2014 turned to a computer program to advise their board on where to invest. Here are two extracts from the article (it’s worth reading the whole thing [see here]):
A Hong Kong venture capitalist fund credits a single member of its management team with pulling it back from the brink of bankruptcy. But the executive is not a seasoned investment professional, nor even a human being. It is an algorithm known as Vital.
Dmitry Kaminskiy, managing partner of Deep Knowledge Ventures, believes that the fund would have gone under without Vital because it would have invested in “overhyped projects.” Vital, which stands for Validating Investment Tool for Advancing Life Sciences, helped the board to make more logical decisions, he said.
While Deep Knowledge Ventures is credited with being the first company to have appointed an AI to its board, Kaminskiy said the algorithm’s role was a little different from that of human directors. “We treat it as a member of our board with observer status,” he said. “As a board, we agreed that we would not make positive investment decisions without corroboration by Vital.”
Not quite the first “real” computer-based director but clearly a key component of the board and its decision making.
It would therefore seem that AI, in some form, can play a part on boards and maybe fill one or more positions on the board, maybe at least in a specialist capacity.
I started to look for views on the feasibility of having an AI-enabled entity as a full member of a board. A Harvard Business Review article suggests that currently AI’s best role is to augment the existing board structure and composition but not replace it.
The article concludes that:
It is only by achieving the optimal balance between humans and machines that Boards can navigate the future of work. Indeed, companies in the new economy will continue to need strong NEDs; albeit with a different mandate, serving a different role, and with different skills. As one senior NED eloquently puts it, “you can’t come up with an algorithm for experience.” Maybe so… just don’t mention that to AlphaGo.
This article (Changes to company law would enable use of ‘robot directors’) on the Board Agenda website looks at the feasibility of making an AI based robot a full and accountable member of the board.
AI in the boardroom
Other proposals have been made to ensure legal protection for directors using AI. Last year Australian lawyer Samar Ashour argued that the problem was the insistence in the law that directors be “natural persons”.
This could be tackled with a change to a company’s constitution and as a result avoiding changes in the law. “If the use of AI is confined to administrative aspects of corporate governance, such as meeting preparations, as opposed to work that encompasses operational and strategic decision-making, the view can be taken that directors are less likely, if not unlikely, to be held accountable by reason of reliance on AI,” Ashour writes.
Others have observed that AI could “augment” almost half of non-executive tasks. Writing in the Harvard Business Review, Ravin Jesuthasan and Shai Ganu of Willis Towers Watson say: “NEDs could likely spend more of their time as internal consultants and advisers to the CEO and management.”
The exploration of AI and how it could be become an effective board members is still under way. There is clearly an interest in business circles. It may just need a political will now to bring about changes in the law and make it happen.
Robots serve humans
“Other legal entities include e.g. a private limited liability company, municipal and other government agencies, etc., all of which have legal personality. They too serve humans. So if a robot is defined as an entity that is demonstrably there to make the life of humans easier, then it would be possible to appoint it as a director under Dutch company law. Only one question then remains: is it desirable?”
There is no unequivocal answer as to whether “it is desirable for a robot to be appointed as a director.” Advantages can certainly be cited, such as increasing the freedom of establishment, but also speeding up decision-making processes. In China, one entity has already appointed a robot as a director for this very reason, and several foreign jurisdictions are considering and have already adopted regulations on robotics and artificial intelligence to some degree.
Conclusion
Well it certainly looks like multiple companies and governments are considering both the feasibility and desirability of allowing an AI Based entity to perform a fully responsible role on a board. There are clearly a lot of regulatory and legal hurdles to be jumped but it does seem that we are moving, albeit quite slowly at the moment, to a point where we could see a board be comprised, at least in some part, by a non-human entity. In a previous blog piece that I wrote about AI my advice was that for all technology-related investments and developments you should consider three questions:
- Is it Possible?
- Is it Feasible?
- Is it Desirable?
Day by day, more and more solutions will become available that will tick the Possible box for a particular problem. The second step is that while it is Possible, is it Feasible for you to implement it? Reasons for something that is Possible but not Feasible are that it would cause too much disruption to the current working practices, it would mean a change in the constitution of the organisation etc. BUT even then when both Possible and Feasible boxes have been ticked the really key point remains is it desirable for you to make this change? How does it feel? Does it feel right? Is it Desirable? If all three boxes are ticked we may have an environment in which we see the first fully legal non-human board director in the UK.
Finally I will give the last word to AI itself. I asked, via ChatGPT, the following questions and here are the answers. You will see my role as a NED may be safe for a little while longer, well that is the view of AI today but then it may have new data to work with tomorrow.
Why can’t an AI robot be a company director?
ChatGPT’s answer:
The main reasons why an AI robot can’t be a company director currently are:
- Legal Framework: Most legal systems don’t recognize AI entities as capable of holding legal responsibilities or rights akin to humans.
- Accountability: AI lacks the ability to truly understand ethical and legal responsibilities or be held accountable for its decisions in the same way humans can.
- Complex Decision-making: While AI can make decisions based on data and algorithms, it lacks human judgment, intuition, and contextual understanding, which are often crucial for effective leadership and decision-making in a corporate setting.
- Ethical Concerns: Appointing an AI as a director raises significant ethical questions about power dynamics, fairness, and the role of humans in governance and leadership positions.
- Public Perception: There may be public resistance or skepticism toward the idea of AI entities holding positions of authority due to concerns about job displacement, loss of human control, and potential negative consequences.
- Until these issues are addressed through comprehensive legal frameworks, technological advancements, and societal consensus, it’s unlikely that AI robots will be appointed as company directors.
Should a robot be a board director?
ChatGPT’s answer:
The decision to appoint a robot as a board director involves careful consideration of legal, ethical, and practical implications. While AI can offer valuable insights and analysis, it lacks human judgment, empathy, and ethical reasoning crucial for effective governance. Additionally, issues of accountability, public perception, and the role of humans in decision-making must be addressed. Ultimately, any decision to involve AI in corporate governance should prioritize transparency, human oversight, and alignment with legal and ethical standards.
–––––––––
Learn more about our secure, user-friendly board portal, Boardlogic here
Or why not schedule a demo with us here
Alan Hewitt