Worldwide spending on artificial intelligence is forecast to reach $632 billion by 2028, yet many companies pioneering the technology are struggling to demonstrate profitability, prompting a reassessment of business models across industries.
The generative AI boom has been fueled by substantial investment, including massive capital expenditure on specialized computing infrastructure, but translating those investments into sustainable earnings remains a significant challenge, according to a recent report by the Harvard Business Review. The report, featuring insights from Harvard Business School faculty member Andy Wu, highlights a disconnect between investment and returns.
Traditionally, companies have relied on established models like Software-as-a-Service (SaaS) to monetize their offerings. However, the unique economics of AI – particularly the variable costs associated with inference and model performance – are forcing product leaders to blend traditional approaches with new strategies, including usage fees and premium add-ons. ProductSchool, in a December 2025 analysis, identified ten emerging AI business models, noting that companies are seeking to “turn intelligence into predictable revenue growth without letting variable compute costs erase your margin.”
A key concept driving this shift is the “AI factory,” where data is ingested, refined, and used to deliver predictions, pattern recognition, and automation. This foundation allows companies to build AI-powered products and services, but too necessitates a rethinking of how value is captured. The World Economic Forum recently outlined the emergence of “AI-first” organizations, designed around AI-native operating models as the primary means of creating and scaling value. These organizations embed intelligence across all workflows and decisions, rather than treating AI as a supplementary layer.
Forbes reported in July 2025 on four emerging AI business models reshaping the enterprise landscape, suggesting a fundamental shift in how businesses operate. These models are not simply about adding AI to existing processes, but about building entirely new mechanisms for value creation and delivery.
The implications extend beyond technology companies. The need to adapt business models is prompting investors to re-evaluate companies across all sectors, as the potential for disruption from AI becomes increasingly apparent. The question of whether profitability ultimately matters, even in the face of massive investment, is a central debate within the industry.
