The Keys to Improving AI’s Transparency and Accountability

Since 2022, when generative Artificial Intelligence (AI) became widely accessible, the world has seen a sharp increase in the utilization of AI in work and daily life, with Statista reporting that AI is now automating more than 60% of global digital activities. While AI is being adopted at scale and scope, one concern remains: the transparency and accountability of AI. Problems exist when some AI models, for example, decline loans for a person from a certain race, give a wrong prediction that results in millions of dollars lost, and spread misinformation. Thus, at the 19th Annual Meeting of the Internet Governance Forum (IGF) in Riyadh, Saudi Arabia, the session "Exploring Transparency and Explainability in AI: An Ethical Imperative" aimed to address the growing concerns surrounding AI governance. As AI continues to influence decision-making across sectors, ensuring that its processes remain transparent and accountable is critical for maintaining public trust.

THE NEED FOR ETHICAL AND TRANSPARENT AI

One of the most thought-provoking notions from the event came from Doreen Bogdan-Martin, Secretary-General of the International Telecommunication Union (ITU), who highlighted the urgent need to address the AI governance gaps. The Secretary-General stressed that ethical AI must be a global priority, as without such, the problems associated with AI biases mentioned would only surface more, for the world’s journey toward AI adoption is ever upward. She also pointed out the issue of inclusive AI—as a third of humanity remains offline without inclusive AI policies, the advancements of AI, which mostly happen in the Global North, risk deepening global disparities and creating a “Digital Divide.” In the same tune as the ITU’s Secretary-General, another panelist, Polish Deputy Prime Minister Dr. Krzysztof Gawkowski, emphasized that AI must reflect values, namely fairness, openness, and respect for human rights. Besides, Dr. Gawkowski doubled down on the primary concern area that without clear regulations, AI could and would perpetuate bias, misinformation, and privacy violations, threatening democratic institutions and citizens’ livelihoods.

ON CREATING A FAIR AND ACCOUNTABLE AI

Panelists provided key recommendations for improving AI transparency and accountability. Ivana Bartoletti, the Chief Privacy Officer (CPO) of Wipro, a global IT solutions company, emphasized the importance of tackling algorithmic bias, particularly in relation to the digital gender gap. The Wipro’s CPO argued that a lack of diversity in AI development risks embedding existing societal inequalities into automated decision-making processes and that the solution for this is simple—a more inclusive training dataset must be the norm in developing a new AI model. In addition to that, Dr. Amal Seghrouchni, Morocco’s Minister of Digital Transition and Administration Reform, chimed in on the creation of ethical AI in practice. The minister highlighted her country’s leadership in implementing UNESCO’s recommendations on AI ethics and called for international cooperation to establish unified AI governance frameworks to ensure fairness and oversight in AI-driven systems.

AI’s long-term success depends on its ability to be transparent, inclusive, and explainable. Therefore, it befalls the government to prioritize privacy protection, algorithmic fairness, and accountability mechanisms that allow individuals to understand how AI decisions are made. Meanwhile, on the corporate front, companies must work with the policymakers to ensure that their AI models meet the standard and share their expertise with the notion that technological advancements serve humanity equitably as the guiding belief. Finally, robust partnerships between governments, businesses, and civil society to foster AI systems that inspire confidence rather than fear must be created and maintained in an increasingly digital world so that technology works for all, not just some.

#BRANDi #BRANDiGlobal #ThinkWithBRANDi #GoodtoGREAT #BiOST #GlobalEconomy #Sustainomy #FutureReady #AI #Transparency #Accountability