Revenge of the AI bubble

The piece highlights recent developments including GitHub's shift of its Copilot product to usage-based billing, which moves away from flat-rate pricing. Additionally, a top executive is quoted as saying that spending on AI is 'getting harder to justify,' suggesting increased scrutiny on returns from AI investments.

The article notes that the AI bubble debate has gone through several phases, and these recent signs indicate a potential cooling of enthusiasm. GitHub's move to usage-based billing means customers pay per use rather than a fixed fee, which could reflect changing demand or a desire to align costs with actual value received. The executive's comment underscores growing pressure on companies to demonstrate tangible outcomes from their AI expenditures.

The article questions whether the value generated by AI will be distributed across the companies that are paying to deploy it, rather than being captured primarily by a few large technology providers. This is a central tension in the current AI landscape, as many firms invest heavily without clear returns. The piece does not take a definitive stance but presents these developments as indicators of a maturing market.