Future of AI: Governance, innovation, and the battle for control

Who gets to define AI’s future? If AI is to serve society as a whole, these fragmented discussions must converge into a more collaborative, interdisciplinary conversation. Policymakers must engage with AI experts to craft informed regulations, businesses should balance profit-driven AI adoption with ethical responsibility, and technical communities should find ways to translate their insights into mainstream discussions.

Artificial intelligence (AI) has sparked both optimism and concern among different groups. While some see it as a game-changing force for good, others worry about its risks. A new study "The hopes and fears of artificial intelligence: a comparative computational discourse analysis" published in AI & Soc (2025) examines how AI discourse is shaped by various actors - politicians, consultancies, and online lay experts - using computational social science methods.

The research by Elmholdt, K.T., Nielsen, J.A., Florczak, C.K. et al. explores how different groups articulate their perspectives on AI, revealing the competing narratives that define its societal impact. Let's take a quick dive into the key findings:

Diverse AI discourses

Politicians tend to frame AI as a societal issue requiring governance, ethical oversight, and regulatory intervention. Their discourse revolves around themes of job displacement, data privacy, and national security. They discuss AI as something that needs to be controlled to maximize its benefits for the public, often highlighting its potential in areas such as healthcare and public services. However, their discussions often lack technical specificity, making it difficult to translate broad ethical concerns into actionable policies. On the other hand, business consultancies take a different approach, portraying AI as a driver of economic transformation. Reports from firms like McKinsey and Accenture focus heavily on automation, digital efficiency, and profitability, viewing AI as an essential tool for companies looking to stay competitive. Unlike politicians, consultancies rarely engage with ethical concerns or AI’s long-term societal consequences, instead positioning it as an inevitable force that businesses must adopt or risk falling behind.

Meanwhile, lay experts on Reddit engage with AI on a technical level, discussing its capabilities, breakthroughs, and limitations in great depth. These discussions focus on neural networks, machine learning models, and AI’s real-world feasibility, offering a mix of enthusiasm and skepticism. Unlike politicians or businesses, Reddit users do not necessarily see AI through a regulatory or financial lens; instead, they dissect its strengths and weaknesses, often debating issues of bias, misinformation, and the limits of AI hype. This creates a space where AI’s progress is critically examined, but their discussions rarely translate into broader policy or business decisions.

Beyond just categorizing discourse, the study also explores the emotional undertones present in these conversations. Politicians are generally the most optimistic about AI’s potential, particularly in governance and healthcare. Consultancies remain neutral, neither overly excited nor fearful, instead emphasizing AI adoption without addressing its risks. Lay experts, however, express a more complex range of emotions, balancing excitement over AI advancements with skepticism about bias, misinformation, and exaggerated claims. While there is significant optimism across all groups, concerns about AI’s ethical implications, its influence on labor markets, and its potential to spread misinformation persist.

AI in Healthcare: A Microcosm of the Larger Debate

One of the most widely discussed applications of AI across all three groups is its role in healthcare. AI’s ability to diagnose diseases, streamline hospital workflows, and enhance predictive medicine generates significant excitement. Many believe that AI could revolutionize the medical field, making healthcare more efficient and accessible. However, concerns about data privacy, algorithmic bias, and over-reliance on AI-driven diagnostics raise important ethical questions. While businesses champion AI’s potential in healthcare, policymakers stress the need for regulations to prevent misuse, and AI enthusiasts debate whether AI models are truly capable of unbiased decision-making in life-or-death scenarios.

Bridging the AI governance gap

The study reveals that AI is not a single, universally understood concept but rather a contested issue shaped by multiple narratives. Politicians focus on AI’s governance challenges, businesses highlight its economic benefits, and lay experts analyze its technical strengths and weaknesses. The problem, however, is that these groups often operate in silos. Policymakers lack technical knowledge, leading to vague or reactionary regulations. Businesses prioritize AI’s profitability, often overlooking ethical concerns. Meanwhile, AI enthusiasts on platforms like Reddit engage in deeply technical discussions but have limited influence on shaping actual AI policy.

This disconnect raises an important question: Who gets to define AI’s future? If AI is to serve society as a whole, these fragmented discussions must converge into a more collaborative, interdisciplinary conversation. Policymakers must engage with AI experts to craft informed regulations, businesses should balance profit-driven AI adoption with ethical responsibility, and technical communities should find ways to translate their insights into mainstream discussions. 

So, based on these findings, it is fair to conclude that AI’s future isn’t just about technological advancements - it’s about who controls the narrative and how we collectively decide to harness its power.