Any guesses which is the biggest enterprise AI company in the world right now? It’s not OpenAI or Anthropic, but it could well be Accenture, ServiceNow, or any other IT or consulting firm.
In FY2024, Accenture raked in $900 million from generative AI alone, up 9x from the previous year, and ended with $3 billion in bookings. This year, it just reported $1.5 billion in generative AI bookings.
ServiceNow, meanwhile, says its AI product Now Assist will hit $1 billion in ACV by 2026. No fluff. Just numbers. Clearly, the direction is clear.
While model labs are burning billions to “solve intelligence,” Accenture and ServiceNow, just to name two, are out here solving invoices. Because someone still needs to build, deploy, and maintain all those intelligent systems.
OpenAI seems to have gotten this lesson. It is no longer content with just powering those systems in the background. It wants to own the last mile.
According to The Information, OpenAI is courting large enterprises and government clients, ramping up its consulting muscle. The AI company achieves this by embedding its own engineers and researchers to tailor AI systems for specific business needs, adopting a strategy similar to that of Palantir, Accenture, and ServiceNow.
The media firm cited insiders and customers saying OpenAI is expanding a services arm where its technical teams help organisations fine-tune models like GPT-4o using their proprietary data. This is something that OpenAI started two years back with Morgan Stanley, building AI bots using their proprietary data.
The goal is to build AI applications, such as ChatGPT-style chatbots, that are tailored for internal workflows and business contexts. This level of support doesn’t come cheap: OpenAI typically asks for a minimum spend of $10 million for these high-touch deployments.
The company had earlier launched a team of Forward Deployed Engineers, who essentially are consultants writing code, helping large enterprises fine-tune models, build apps, and integrate AI into their own workflows.
OpenAI Aims to Compete with the Consulting Giants
This enables OpenAI to compete directly with consulting firms like Accenture, Deloitte, EY and Palantir.
Even Thinking Machines Lab, founded by OpenAI’s former CTO Mira Murati, according to recent reports, is also working on AI for Enterprise. The same applies to Anthropic, which generates a significant portion of its revenue through API usage and enterprise offerings.
What we’re seeing here isn’t just a consulting land grab. It’s a complete redefinition of what enterprise AI even means. But why is this even happening?
Saanya Ojha, partner at Bain Capital Ventures, said in a post that the dream of AI-native software—where GPT writes your CRM and agents automate away your workflows—sounds fantastic in theory. But “who maintains the thing? Who integrates it into your stack? Who trains the team, manages compliance, handles edge cases?” Ojha said, adding that it would not be free.
Welcome to the new world where software is the easy part. Ownership is the real moat. This is where consulting firms shine and startups start to look more like service companies. Because AI isn’t a one-click SaaS product. It’s an ongoing engagement. It’s software with a long tail.
If OpenAI’s move into services sounded like a bold pivot, industry insiders see it as inevitable. Kabir Choudhury, associate consultant at EY India, said, “It’s not just about models, it’s about deployment as differentiation.” The moat isn’t in how powerful your model is, it’s in how well you make it work within an enterprise’s chaos.
Israel Wagshul, Associate General Counsel at Similarweb drives the point home for OpenAI. “Enterprise AI is still a people-heavy transformation game,” he said. Custom software still breaks, still costs a fortune, and still resists scaling. OpenAI is trying to scale both the model and the services layer is, as he puts it, “great on paper, but risky in practice.”
All said, the shift is clear. Deployment is the moat. And OpenAI just walked right into the battlefield.
Thiyagarajan M from Upekkha, commented on Ojha’s blog that, “OpenAI hiring consultants isn’t about services, it’s admitting that in enterprise, distribution beats product.”
Everyone’s a Consultant Now
OpenAI may sell APIs, but what it’s really pushing is transformation-as-a-service. And the distinction between product and service is vanishing fast.
The recent example of OpenAI partnering with the Indian IT giant HCLTech is one such example. It might help HCLTech and its clients test and deploy OpenAI products, directly in sync with the IT giant’s clients to figure out what works and what doesn’t.
The startups need to realise this hard truth. Enterprises don’t want DIY APIs, instead, they look for measurable outcomes. And until AI gets good enough to plug-and-play into SAP or Salesforce, someone has to do the plumbing.
Even Palantir, once allergic to being called a services company, now proudly runs on enterprise contracts that look suspiciously like system integrator deals.
The harsh truth for AI startups is that selling a devtool, or worse, an AI widget, isn’t cutting it. Enterprises want end-to-end solutions, not cool demos. They want partners, not platforms.
Is Indian IT Playing the Right Cards?
Perhaps, Indian IT is already on the path that Accenture and ServiceNow are taking, and now even OpenAI is following. Although TCS, Infosys, Wipro, and others refuse to report revenue from generative AI, it might be the most significant driver for them at this point.
While Accenture booked $1.5 billion in AI revenue in Q3 FY25 alone, Indian IT is still experimenting with pilot projects and “AI-first” decks.
TCS says AI is in “every conversation,” but not in any revenue disclosure. Infosys has over 225 GenAI projects, but lacks clarity on their impact. Wipro closed 17 AI-infused deals in Q4 FY25, with the guidance reports forecasting a revenue decline for the subsequent quarters. This needs to change swiftly, which might be visible in the earnings call starting next week.
Read: Indian IT Loves AI Washing
The post Accenture, ServiceNow and Palantir Teach Startups How to Make AI Money appeared first on Analytics India Magazine.
