When Bradford Newman began advocating for more artificial intelligence skills in the C-Suite in 2015, “people were laughing at me,” he said.
Newman, who heads the machine learning and AI practice for global law firm Baker Mackenzie at his Palo Alto office, adds that when he mentions the need for companies to appoint a chief AI officer, people usually respond, “What is it?”
But with the use of artificial intelligence pervasive across the enterprise, and issues surrounding AI ethics, bias, risk, regulation and law currently circulating across the business landscape, the importance of appointing a chief AI officer is clearer than ever, he said.
This belief led to a new Baker Mackenzie report, released in March, called “Risky Business: Identifying Blind Spots in Corporate Oversight of Artificial Intelligence.” The report surveyed 500 US-based, C-level executives who identified themselves as part of a decision-making team responsible for the adoption, use and management of their organization of AI-enabled tools.
In a press release on the publication of the survey, Newman said: “Given the increase in state law and regulatory enforcement, companies need to increase their game when it comes to AI oversight and governance so that their AI is ethical and protects itself from liability. Manage their risk exposure accordingly. ”
Corporate blind spots about AI risk
According to Newman, the survey found significant corporate blind spots around AI risk. For one thing, C-level executives increased the risk of AI cyber intrusions but minimized AI risks related to algorithm bias and reputation. And while all the executives surveyed said their board of directors had little awareness of the potential enterprise risk of AI, only 4% rated the risks as “significant.” And more than half described the risk as ‘somewhat significant’.
The survey also found that organizations “lack a firm grip on bias management once AI-enabled tools are installed.” When managing implicit bias in AI tools in-house, for example, only 61% have a team for up-rank or down-rank data, while 50% say they can override some – not all – AI-enabled results. Is.
In addition, the survey found that two-thirds of companies do not have a Chief Artificial Intelligence Officer, which makes AI surveillance fall under the domain of the CTO or CIO. At the same time, only 41% of corporate boards specialize in AI on them.
AI Regulation Inflation Point
Newman insisted that more attention be paid to AI in the C-suite and especially in the boardroom.
“We are at a turning point where Europe and the US are going to regulate AI,” he said. “I think corporations are reacting badly on their hind legs, because they don’t get it – they have a false sense of security.”
While it is anti-regulatory in many areas, Newman claims that AI is very different. “There should be an asterisk through it because of the effect of AI,” he said. “It’s not just about computer science, it’s about human ethics … it goes to the essence of who we are as human beings and the fact that we are a Western liberal democratic society with a strong view of individual rights.”
From a corporate governance perspective, AI is also different, he added: “For example, unlike financial work, which is accounted for in dollars and cents within the corporate structure and is properly reported and disclosed to our shareholders, Artificial Intelligence and Data. Science includes law, human resources, and ethics, “he said.” There are many examples of things that are legally permissible, but not compatible with corporate culture. ”
However, AI tends to be fragmented and diverse in the enterprise, he explained.
“There is no one-size-fits-all regulation where the person who means good can go to the C-suit and say, ‘We need to follow this. We need to train. We need compliance. ‘ So, it’s still a kind of theoretical, and sea-suits don’t usually respond to the theoretical, “he said.
Finally, Newman added, there are many internal political elements around AI, including AI, data science, and the supply chain. “They all say, ‘It’s mine,'” he said.
Requirement of Chief AI Officer
Newman said the Chief AI Officer (CAIO) – that is, what would help in the appointment of a C-suite level executive who reports to the CEO at the same level as the CIO, CISO or CFO. The CAIO will have the ultimate responsibility for overseeing all AI matters in the corporation.
“A lot of people want to know how a person can fit into that role, but we’re not saying that CFO knows every calculation of the deep-rooted financial aspects of a corporation – but it does report it,” he said.
The CAIO will therefore be charged with informing shareholders and external regulators and governing bodies.
“Most importantly, they will play a role in corporate governance, oversight, oversight and all things AI compliance,” Newman added.
However, Newman acknowledges that the idea of installing CAIO will not solve every AI-related challenge.
“Will it be perfect? No, nothing – but it will be a big step forward,” he said.
The Chief AI Officer should have a background in some aspects of AI, computer science, as well as some aspects of ethics and law.
While one-third of Baker McKenzie’s surveys say he currently has a “like” chief artificial intelligence officer, Newman believes those are “liberal” figures.
“I think most of the boards are far behind, relying on the patchwork of the chief information officers, the chief security officers or the HR heads sitting in the C-suite,” he said. “It’s very tangled together and what I’m talking about as a real CAIO is not a true job description kept by a person with the kind of oversight and metrics responsibility.”
The future of the Chief AI officer
These days, Newman says people no longer ask, ‘What is a Chief AI Officer?’ As much as But instead, organizations claim they are “ethical” and their AI is not implicitly biased.
“There is a growing awareness that corporations need to be monitored, as well as a misunderstanding of security that the oversight that currently exists in most organizations is adequate,” he continued. “It’s not enough when regulators, enforcers and plaintiffs come – if I change sides and start representing clients and plaintiffs, I can make huge holes in most corporate oversight and governance. For AI. ”
Organizations need a chief AI officer, because “the questions asked by this technology are more than zero, issues, data sets.”
Organizations are “playing with live ammunition,” he said. “AI is not an area that should be left to the data scientist alone.”