Editors of TechCrunch in the Equity podcast suggested that GitHub Copilot’s move to price-through-token could trigger a chain reaction of price revisions across the AI product industry. Shifting a portion of users from fixed subscription to payment for the volume of information being processed is effectively signaling a paradigm shift: service costs are increasingly linked to real consumption rather than abstract access.

Experts link this shift to the fundamental economics of big models: high infrastructure costs and incentives are forcing companies to seek more sustainable monetization models. Anthony Ha noted that the market is still heavily subsidized by venture capital, but as they prepare for an IPO, players like Anthropic will have to demonstrate a path toward profitability - which almost inevitably means shifting some of the costs onto users. Sean O'Kane added that even tech giants like Uber are already running out of AI tool budgets and having to impose internal limits, which calls into question the developers' ability to reduce technology costs quickly enough.

In parallel, the regulatory context is changing: on June 2, 2026, the US president signed a decree allowing authorities to test powerful AI models before public release. Kirsten Korosek emphasized that the trend of "tokenmaking" had taken place in a matter of months and was already causing market irritation, but would probably become a new norm. Against this background, Gartner’s forecast that global AI spending will rise to $2.59 trillion in 2026 looks not like a guarantee of expansion, but rather as a challenge: whether the token economy can scale without a massive exodus of users.