For two years, enterprise AI ran on one question: how fast can we spend? Bigger context windows, more tokens, more agents, more pilots. The size of the invoice was treated as proof of ambition. In June 2026 the question changed, and the new one is less comfortable: what did any of it return?

I have spent this year deploying AI inside a large enterprise, in the rooms where these budgets are defended. What this month’s news points to is plain enough. AI as a science experiment is ending, and AI as a line on the P&L has begun. Organisations that got value will be easy to tell apart from the ones that only got invoices.

The tell: from "tokenmaxxing" to efficiency

The clearest signal came from the demand side. Reporting late in the month described enterprise users shifting from "tokenmaxxing" (maximising usage at any cost) toward cost efficiency, just as OpenAI and Anthropic, the chief beneficiaries of the spend-at-all-costs mindset, gear up for trillion-dollar IPOs. In that coverage, the chief executive of one AI startup switched his company entirely off premium frontier models, moved 100% of its traffic to a cheaper open-weight alternative, and watched his cost curve, in his words, crash to the ground.

One anecdote is not a trend. But it fits everything else that happened this month. The buyer has stopped asking "how powerful is it?" and started asking "what does it cost me per outcome, and can I get the same outcome cheaper?" That is more than a change in tone. It is the point where a technology stops being a frontier and becomes a budget line.

The CFO has entered the chat

The people now shaping AI strategy are not the ones you would expect. Across industries, chief financial officers are moving aggressively to impose budget controls on AI projects, replacing the open-ended experimentation of the last two years. The numbers explain why. As enterprise AI moves from pilot to production, only about one in four AI initiatives is delivering the ROI that was promised, even as vendors cite headline returns of several rupees for every one invested. Gartner now expects more than 40% of agentic AI projects to be cancelled by 2027, and reports that only a fifth of companies running autonomous agents have a mature governance model for them.

Put those two facts together and the cancellations are easy to explain. The shortfall is one of discipline. Thousands of pilots were funded on what AI might do, with no gate for what it delivered and no control plane to govern it once it did. That bill is now coming due.

Why the cost curve is finally bending

The supply side is responding to the same pressure, which is good news for buyers. The main infrastructure story of the month was vertical integration aimed at the cost of inference. OpenAI unveiled its first custom inference chip, Jalapeño, built with Broadcom, to reduce its structural dependence on Nvidia GPUs. Microsoft pushed its own in-house MAI models, explicitly designed to lessen reliance on OpenAI and lower costs for developers. And cheaper open-weight models, often Chinese and MIT-licensed, kept pulling down the floor price of "good enough" intelligence every month.

I have argued this all year, and the hardware now backs it up: the model is a commoditising input. When the most powerful labs are racing to make inference cheaper and the open-weight tier is closing the quality gap, the rational enterprise posture is portability. It lets you ride the cost curve down without re-platforming every time a cheaper option appears. Loyalty to a single supplier does not.

But cheaper compute is not the same as ROI

Here is the trap I would warn every leader about. Cutting your cost per token is not the same as creating value. You can make a worthless pilot 90% cheaper and it is still worthless. An efficiency drive can optimise the wrong number and celebrate savings on initiatives that should never have been scaled.

The organisations getting returns are doing something quieter and harder. They govern. When Microsoft and KPMG scaled agents across a 276,000-person workforce in June, the headline word was "governed", not "powerful": centralised control of agents operating across systems, data and processes. The most valuable agents this year are the unglamorous ones running autonomously inside companies, where a retailer can quietly let an agent negotiate supplier contracts. The demo reels get the attention. The value is coming from governed, narrowly scoped deployment, and spend alone does not produce it. Neither do savings.

What this means for enterprise leaders

I am telling my own organisation to treat this month as the end of the experimentation budget and the start of the accountability budget. That comes down to four commitments. First, treat the model as a swappable, commoditising input and build for portability, so falling prices flow to your margin instead of locking you in. Second, put a CFO-grade ROI gate in front of every agent before it scales: a defined outcome, a baseline and a number. An agent with no measured return is a liability with an API key. Third, fund the governance and control plane first, since ungoverned sprawl is the reason 40% of these projects will be cancelled. Fourth, measure the quiet wins, not the demos, and reward the teams shipping dependable, governed automation over the ones with the best prototype.

India has an advantage in this. The hype era rewarded whoever could spend the most, and that was never going to be us. The efficiency era rewards cost-aware, domain-wrapped engineering, which Indian technology firms have spent decades building. As the model layer commoditises and the global question becomes "what is the governed, ROI-positive outcome?", the advantage moves from the firms with the biggest compute budgets to the firms with the deepest domain discipline.

My take

The hype era funded the experiments. The efficiency era will decide who got value from them, and the verdict will be harsher than the spending implied. The winners of 2027 will be the organisations that started governing AI like a P&L instead of a science project. That means portable at the model layer, disciplined at the ROI gate, serious about governance and honest about which pilots deserved to die. The bill has come due. Paying it properly is also how you start earning a return.