Merelbeke, 22 July 2020

Belgian business leaders are calling for the creation of a regulatory framework to ensure the ethical use of artificial intelligence (AI). According to research by Avanade, the leading digital innovator on the Microsoft ecosystem, the majority of C-suite executives say their companies do not currently have a policy in place to ensure ethical responsibility for their AI-tools. Additionally, one third of respondents admits that their AI-tools have most likely made unethical decisions in the past. The study reveals that most businesses are doing very little to guarantee ethical responsibility in their AI-applications and do not know where to turn to for more support.

Johann Corlemeijer, Country Manager at Avanade Belgium, explains: “As a result of COVID-19, we are facing an expedited digital revolution in which AI will most certainly play a significant part. Recent exposures of institutional racism prove all the more that business leaders need to ensure systemic biases are not present in technology. Because regulation often lags behind technological advancement, we need to be aware of the risks involved and prioritize digital ethics.”

Most concerns about security, privacy and liability

The research shows that Belgian business leaders are mostly concerned about security breaches (84%), invasion of privacy (82%) and liability issues (74%). Patrick Tack, Data & AI Lead at Avanade Belgium, is not surprised that executives are most concerned about privacy and security breaches: “We are encouraged by the concern about security and privacy issues. However, digital ethics failures have the potential to have a much more devastating impact on people than security and privacy breaches. For example, we have already seen systemic bias in health care, financial services, and recruitment processes. Outcomes like someone not getting a job because of their gender due to a mistaken algorithm should get at least as much of our attention as the possibility of privacy breach or identity theft”. Luckily, the research shows that 7 out of 10 respondents are concerned about societal bias being present in AI-systems.

Uncertainty and lack of oversight

The survey results show that Belgian business leaders are not well informed about the ethical risks related to artificial intelligence. Nearly half (44%) of the respondents are unaware of the risk of systemic bias in AI-systems, and one in three (33%) are unaware of transparency issues, such as a person being unaware that they are interacting with an AI-based tool. Moreover, one in five (22%) are unaware of privacy risks, such as AI-systems that collect and use private data without the knowledge or consent of the individual.

Consequently, only a minority of the respondents admits covering the basics to ensure ethical responsibility in their AI-tools. More than two thirds (70%) of the respondents admit that they do not have a policy in place to ensure ethical responsibility in their AI-tools. So, unsurprisingly, one-third (33%) of the respondents who use AI admit that there is a high probability that their AI-tools have made an unethical decision since their implementation. Additionally, half of Belgium’s executives (52%) believes that they will be unethically treated by an AI-tool themselves.

“These figures show the uncertainty that is prevalent among Belgian business leaders. There is an imminent lack of oversight,” Patrick Tack states, “Artificial intelligence is an incredible kind of technology that will propel numerous industries forward. However, it is crucial that business leaders and their organization are educated on the ethical implementation and use of artificial intelligence. This responsibility lies with the developers, not the end-user.”

Need for regulation

Four out of five respondents (79%) cited the ethical use of artificial intelligence as a priority for their business in the next three years. The survey indicated a strong consensus amongst Belgian business leaders, as nine out of ten (88%) state that the ethical aspect of artificial intelligence needs to be regulated.

Johann Corlemeijer, Country Manager at Avanade Belgium, continues: “When faced with a global crisis, the importance of transparency and trust is crucial. People are dealing with enough pressure in their own lives that they need interactions with their chosen business to be supportive and genuine. As more and more Belgian organizations adopt AI-technology, there is a burning need for a regulatory framework to ensure we get the full benefits of artificial intelligence while addressing the key ethical implications. In the meantime, business leaders have to make ethically minded decisions based on corporate values and guiding principles. Artificial intelligence should be trustworthy, ethically and socially responsible, and respecting of our individual rights.”