Challenges in AI product development and adoption

Information service providers face significant hurdles when transitioning from labor-intensive models to AI-driven product development. The Artificial Intelligence Foundation (AIF) has highlighted difficulties in identifying viable markets, establishing sustainable pricing, and managing the high costs of Proof of Concept (POC) projects, which clients frequently treat as free demonstrations.

As the technology evolves, experts warn of a “build vs. buy” trap. While generative AI has lowered the barrier to creating functional prototypes and minimum viable products, the long-term costs of internal development—including security, compliance, integration, and maintenance—are often underestimated.

To achieve successful transformation, industry experts suggest that companies must shift from project-based work to permanent product ownership. This requires dedicated teams for continuous improvement rather than temporary teams that simply deliver a project. Additionally, there is a call for governments to stimulate market demand by integrating AI into public sector domains like infrastructure and transportation to help companies validate their technologies.