Welcome to AI Insights. Today we’re looking at why enterprise AI adoption is stalling. According to market watcher Canalys, businesses poured $90.9 billion into cloud infrastructure and platform services in Q1—a 21 percent jump year-on-year—as they explore generative AI. Yet many are hitting a roadblock: unpredictable inferencing costs.

“Unlike training, which is a one-time investment, inference represents a recurring operational cost,” says Canalys senior director Rachel Brindley, “making it a critical constraint on the path to AI commercialization.” Researcher Yi Zhang adds that usage-based pricing—charging per token or API call—means “when inference costs are volatile or excessively high, enterprises are forced to restrict usage, reduce model complexity, or limit deployment to high-value scenarios.”

After some firms saw runaway cloud bills—like Basecamp’s $3 million annual charge—many are reevaluating public clouds for AI. Canalys chief analyst Alastair Edwards warns large-scale inferencing in the public cloud “becomes unsustainable from a cost perspective.” As a result, businesses are exploring on-premises, colocation and specialized hosting to tame those bills and fully unlock AI’s potential.
Link to Article