Every CIO has had the conversation by now. A business leader has watched an AI coding assistant spit out a working prototype in an afternoon and asks the obvious question: why are we paying for a platform when we could just build it?
Manish Gupta has heard that question more than most. As Managing Director and Country Head of Liferay in India, he has spent 16 years selling digital experience platforms (DXPs) to banks, ministries, insurers and manufacturers, and he says the AI-will-replace-the-platform argument is gaining airtime inside his customers’ boardrooms.
His answer is blunt, and it is aimed squarely at the CIO’s chair. “Some people see AI as a threat. They feel that using AI, they can create a similar tool like Liferay, and they will not require Liferay,” Gupta says. “But we do not think in that way.”
Buy the plumbing, build the business logic
Gupta’s core argument is about ownership. Code that an AI generates today still has to be tested, secured, upgraded and understood by someone, indefinitely. Liferay’s platform, he points out, has been under development for 15 to 20 years.
“You want to rely on a code generated by an AI tool, or you want to rely on a code which has been developed over the last 15, 20 years, and actively being maintained by someone?” he asks. “If you are creating a similar kind of code base, first of all, you don’t know how stable that code is, how optimized that code is, and how much you can rely on that. Second, you have to continuously maintain it. You have to continuously upgrade it.”
He frames the choice as an identity question for the enterprise itself. “Whether you are into the business of writing the code or you are into the business of selling the insurance or onboarding a banking customer, the moment you start getting into code, you become an IT services company,” Gupta says. “Do you really want to be an IT services company?”
Then comes the argument that tends to land hardest with technology leaders who have lived through the attrition cycles of Indian IT: key-person risk. “Once they write the code and once somebody understands it, what if that person is leaving?” Gupta says. “Who is going to manage it?”
His thoughts line up with the anxieties CIOs themselves are reporting. The Thoughtworks Global CIO Survey 2026 found that 88% of Indian respondents agree AI adoption is moving faster than governance structures can keep up, and that 43% say they are personally accountable for data privacy breaches they cannot fully influence. ServiceNow’s Enterprise AI Maturity Index 2026 reported that AI investment in India grew 119% in a year, yet only 11% of organisations had moved to autonomous workflows. Spending is outrunning control.
Governed agents, not free-roaming ones
Liferay’s own answer to that control gap is AI Hub, a standalone SaaS product the company launched in public beta in June 2026. It is a low-code environment for building, deploying and managing AI agents. According to the company, AI Hub agents act on behalf of authenticated DXP users and can reach only the data that user is already permitted to see, and the architecture is model-agnostic.
Gupta is careful about what Liferay is and is not. “We are not providing any LLM. We are not a frontline model company,” he says. The value, in his telling, is in what surrounds the model. Liferay’s platform is open source and supports connections to AI tools through the Model Context Protocol (MCP), so customers can plug in the AI of their choice. The permissions model does the policing.
“We are using AI within the guardrails. We are not just doing vibe coding. We are doing vibe coding within our framework, so that we know what the AI is generating, we can easily maintain it, we can easily understand it,” Gupta says. “If there is a bug, we can easily fix it.”
The permissioning point is the one he returns to most. “AI cannot just go and access any sort of data. This is one of the big advantages of our AI Hub, where you are deploying the agent which is working within the permissions of our DXP framework, rather than going and accessing any kind of data and creating havoc for the automation.”
For CIOs weighing agent pilots against audit and compliance exposure, that is the pitch in one sentence: inherit the access controls you already trust, rather than building a new governance layer for every agent.
From two products to seven
The AI story is also an internal one for Liferay. A year ago, Gupta says, the company sold essentially two products: its DXP and Analytics Cloud. Today there are seven.
“We have created something called Liferay Data Platform. We have created Content Marketing Platform. We have created Digital Sales Room. We have created AI Hub,” he says. “We believe in eating our own dog food. Whatever we are creating, we are using our own platform to create those tools, and AI has really helped us, because it is amplifying our productivity.”
The product cadence is visible in the market. Liferay announced general availability of its Data Platform on 19 August and of its Digital Sales Room on 9 September, and it has also released a Content Marketing Platform as an add-on to its CMS.
If a company can ship that many products in a year on its own framework, he suggests, a customer can compress its own launch cycles the same way.
He points to one insurance customer as evidence. “Earlier they were launching any new product, it was taking months,” Gupta says. “Now, with Liferay and AI integrations, they are able to do that launch in weeks.” That customer, IndusInd Nippon Life Insurance, was among the organisations presenting at the company’s Customer Summit in Mumbai on 21 October, where its CTO and marketing head discussed accelerating product launches through vertical platforms.
Gupta is careful not to oversell what AI has done for his own company’s market. “It is not that AI has created a lot of opportunity for us,” he says, “but it has really created a lot of additional value for our customers.”
The data problem behind the personalisation promise
If AI Hub is the governance story, the Liferay Data Platform (LDP) is the plumbing story, and Gupta calls it one of the two products with “very huge potential.”
The problem it targets will be familiar to any CIO who has tried to personalise a customer journey. “Every organisation has the data being stored in multiple tools, in multiple repositories, under multiple applications,” Gupta says. “There is no single source of truth.”
He walks through a typical scenario. A marketing team runs a campaign. A CRM team records the outcomes. A website serves the visitor. None of them talk to each other in real time. “If your website doesn’t know that this is an existing customer or this is a new lead, you will not be able to engage with that person coming without that information,” he says.
Liferay says LDP pulls account and customer data from CRM, marketing automation and behavioural sources into one view, resolves conflicts using customer-defined rules, and lets marketing and sales teams build real-time segments that trigger personalised content inside the DXP.
Gupta is candid that this was a gap Liferay itself had been papering over. Customers were already bolting on third-party campaign and data tools. “The integration which was happening was very loose integration,” he says. “There was not a very tight integration that you can really leverage that data best.”
His claim for the new products is that tight coupling lets the DXP use that data to the fullest.
He is also quick to distinguish the offering from the crowded customer data platform category. “CDP is a very well-known product in the market. But how are you creating a CDP which is more relevant for a DXP kind of platform? That’s why we are calling it Liferay Data Platform rather than Liferay CDP.”
The marketing budget is the prize
The commercial logic behind the expansion is less about technology than about whose budget Liferay can reach. Historically, Gupta says, Liferay sold to CIOs and CTOs, and the website was the marketing team’s only touchpoint with the product. That left money on the table. “Marketing is having big budgets in every organisation, and we were not able to really give them a good set of tools to work with,” he says. “Website was just one use case for them, and website used to be just a very small fraction of their overall marketing budget.”
The new portfolio changes the buyer map. “Earlier we were able to basically target CIOs and maybe CTOs. We were never able to really tap the marketing team, or the data team,” Gupta says. “With these products, we now have a whole set of categories where we can target, and not only that, the developer community as well.”
For CIOs, that carries a practical implication. The platform they already run for portals and websites may soon be pitched to their CMO and head of data as well, and the technology leader will be expected to referee the integration, security and governance questions that follow.
A horizontal bet on a public-sector base
Liferay’s India footprint is unusually weighted toward institutions where trust and auditability matter most. Gupta says the company landed State Bank of India, Federal Bank and Apollo Tyres in its early years after starting India operations in 2010, and that eight of India’s 12 public sector banks now use the product, along with several important ministries of the Government of India.
He attributes the public-sector traction to two things: open source and deployment flexibility. “We are an open source company. Whatever we code, anybody can go and see the source code,” Gupta says. On deployment: “You want to deploy it on-prem, you want to deploy it in your data centre, you want to go on cloud, SaaS, all the deployment options.”
That flexibility is also what he credits for the platform’s horizontal reach. Gupta uses an insurer as the worked example: a public website to attract leads, a buying journey to convert them, a customer portal to serve them, an agent portal for the distribution network and partner portals for bancassurance channels.
“All four or five cases where you can acquire the customer, all the use cases you can build on one single platform,” he says. “Irrespective of the database, irrespective of the backend. We are supporting all sort of operating systems, all sort of databases, all sort of deployment.”
Banking, financial services and insurance remain the strongest verticals, he says, because user experience is where those businesses compete. “They were always B2C kind of businesses, so that is where they really wanted to have a good user interface, good customer experience, good post-sales experience,” Gupta says. But he sees manufacturing and pharmaceuticals moving the same way, as they look to offer customers more intuitive interfaces.
Doubling down, then fixing what is built
The India plan for the next one to two years is ambitious. Gupta wants to double the business, and he expects the expanded product line to do most of the work.
“We will not be surprised if the number of opportunities gets doubled, because there is a whole lot of new market that will open up,” he says. “Whatever business we are doing right now, we have the capacity to double it in the next one or two years.”
The first target is the installed base. Many customers, he says, have been asking what else they can do with Liferay, and when they needed campaign or data tools, partners went and integrated third-party products. “All these places where there are third-party tools integrated with Liferay, we have an opportunity to be the first choice for those customers,” Gupta says. “Penetrating within the existing customer base is very easy.”
The most interesting commitment may be what Liferay India says it will not do. After a year of rapid product launches, the company is pausing the expansion. “Next year, we will not be focusing on creating more tools,” Gupta says, “but on how we can enhance the five tools that we have created in the last one year.”