AI pilot era is ending. Now worry about the bill

Meanwhile, the Economic Times daily newspaper is available online now.

Key points

• Enterprise AISnowflakeAI ROIAI adoptionAI inference coststoken costsdynamic model routingCortex AI GatewayVijayant RaiRangarajan Srirangam.
• Rai remarked Snowflake is now seeing "some really large scale adoption" throughout its global and Indian customer base.
• Power shift, a déjà vu: Air India could be new CEO's toughest flight yet.
• As firms deploy AI throughout more users and workloads, Snowflake notes attention is shifting to token costs, model choice, spending controls and returns.
• The piece which obviously now worries most customers is the cost side of things.

Meanwhile, the AI pilot era is ending. Now enterprises are worrying regarding the bill, states Snowflake.

Meanwhile, the AI pilot era is ending. Now enterprises are worrying regarding the bill, states Snowflake. ET OnlineLast Updated: Aug 21, 2026, 10: 08: 00 AM IST.

As AI moves from pilots to production, Indian enterprises are beginning to ask a familiar question: "Kitna deti hai?" Only this time, it is regarding how much business value they are getting for every rupee spent on AI. Snowflake is seeing firms pay much closer attention to inference costs, token consumption and returns on AI investments as projects move beyond pilots and are deployed throughout more employees, customers and business processes, according to Vijayant Rai, Managing Director, Snowflake India. "The piece which obviously now worries most customers is the cost side of things. How plenty of tokens am I going to utilise for a particular apply case? Is it viable? The budgets for it? We are seeing that customers are getting more conscious of that, " Rai informed the Economic Times Digital in an interview.

India a key expansion engine for Snowflake: CEO Sridhar Ramaswamy In the last few years, enterprises have moved from the initial excitement around generative AI to experiments and more lately to larger production deployments. Live Events.

Rai remarked Snowflake is now seeing "some really large scale adoption" throughout its global and Indian customer base. Snowflake has around 13, 900 customers globally, with Rai saying regarding 13, 600 were already using AI features on its platform. As adoption widens, AI economics become harder to ignore, nevertheless. "In most cases, there have been sporadic employ cases that have gone into production. As long as they are giving sales some advantage, whether through customer adoption or acquiring new customers, it's a fairly straightforward conversation, " Rai stated. "But the moment the scale becomes bigger, you've got to start thinking about the cost side of things, " he continued. Cost of using the best model for everything.

That shift is additionally behind Snowflake's latest push around dynamic model routing. Earlier this week, the firm unveiled dynamic model routing within Cortex AI Gateway. It can select an AI model for a particular task based on factors such as cost and performance. The feature is additionally being integrated into Snowflake CoCo, its coding agent, and CoWork, its agent for business users.: India's AI edge will come from talent, not compute, states Snowflake CEO "You might not want to use a frontier LLM for a use case that is simple, large-scale and repetitive, " Rai noted. "You might want to use a lower-cost model that can still do the job, " he continued. Snowflake is additionally expanding access to open models, including DeepSeek-V4-Flash 0731 and GLM-5.3, alongside models from providers such as OpenAI, Anthropic, Google, Meta and Mistral. Rangarajan Srirangam, Senior Regional Vice President of Solution Engineering, Snowflake India remarked model routing is intended to provide customers another option rather than take model choice away from them. "If you strongly feel that one particular model is the best choice, go for it. But if you think the system can create a better decision, you can let it choose, "Srirangam informed the Economic Times Digital. Though Srirangam noted those gains should not be read as equivalent cost savings for every enterprise, snowflake remarked its internal tests demonstrated dynamic model routing improved token efficiency by up to three times in one workload and by 25% in another. "What you really need to do is take a representative set of use cases, maybe your top 50, run them with and without model routing, compare the token costs, and then make a decision, " he remarked. AI's spending difficulty.

As AI moves deeper into day-to-day operations, enterprises are beginning to treat spending controls as part of the deployment itself instead of waiting for the costs to rise. Rai remarked businesses are taking different approaches to funding AI. While others are adding fresh spending.: Enterprises using AI as returns outweigh costs: Snowflake executive The willingness to spend, even so, additionally depends on whether businesses can show a return, some are carving out funds from existing technology budgets. Rai remarked a recent Snowflake survey of around 2, 000 customers globally identified that 71% of Indian respondents documented positive return on investment from AI, compared with 61% globally. "As long as there is ROI, I think the cost is justified because it helps an organisation get ahead in different parts of its business, " Rai remarked. But the equation becomes more complicated as AI is opened up to more employees, applications and autonomous agents. Enterprises then need to know not just how much they are spending overall, but which users, models and workloads are driving that consumption. "For us, a lot of this cost conversation is leading to controls on cost, " Srirangam remarked. "At what granularity can you control costs and what limits can you set and what kind of visibility can you get to spending?" He remarked enterprises are increasingly looking at controls such as per-user and service-level limits, budgets, usage tracking and auditing. According to Srirangam, there is additionally the risk that a simple request can trigger significant computation behind the scenes. "Then there are other things to worry about, such as runaway spending, " Srirangam continued. Education, silos and context.

Cost, nevertheless, is only one reason AI projects struggle to scale. Srirangam remarked the first barrier he sees is often not compute or access to models, but education. "I think the primary hurdle is education. Individuals need to understand how to apply AI and how to obtain the most value from it, " he remarked. Beyond that, Srirangam remarked enterprises continue to struggle with fragmented data and inconsistent business definitions.: Timing, not talent, holding back Snowflake's India R&D expansion: Cofounder Benoît Dageville "If you get past education, the next two problems are silos and lack of context, " he stated. While finance may consider the same accord closed only once payment or consumption starts, rai offered the example of a sales team considering a agreement closed when a contract is signed. "If you do not have the semantic sort of clearly defined, the context will go wrong completely, " he remarked. Add Now!

ET Prime special series: Bigger or Better? Bharat Electronics' 2013 to 2026 journey – Part 2.

For now, AI pilot era is ending. Now worry about the bill remains the part of the story worth watching, and further updates are likely as more details are confirmed.