At MachineCon GCC 2025 Genpact set out a four‑fold playbook and launched an AI Gigafactory and an agentic AP Suite to move pilots into production — but talent shortages, governance questions and weak third‑party validation mean many GCCs still face a hard road from experiment to enterprise.
Global capability centres are shifting from cost centres to engines of innovation — and Genpact is positioning itself as both architect and enabler of that transition. Speaking at MachineCon GCC 2025 in Goa, Harpreet Duggal, Senior Advisor at Genpact, framed the moment as one in which generative and agentic forms of AI are turning lofty digital strategies into operational realities. “With our heritage as a GCC pioneer that brings together decades of deep domain expertise and advanced technology capabilities, Genpact strives to be at the forefront of AI and agentic‑AI‑led change,” he told delegates.
Yet the ambition on display masks a persistent implementation gap. Industry research has for some time warned that many GCCs remain trapped between experiment and enterprise‑grade delivery: legacy organisational structures, thin investment models, scarcity of high‑quality AI talent and weak enterprise orchestration all conspire to slow scale‑up. HFS Research, in a point of view published in October 2024, described precisely this evolution — a move from “cost arbitrage” to “skills and innovation arbitrage” — while stressing the frictions that prevent proofs of concept becoming standardised services.
How, then, do GCCs cross that valley of death? Genpact offers a four‑fold answer that blends capability building and governance, and its recent product and platform announcements show how it plans to operationalise that approach. The company says it provides innovation enablement to help centres adopt emerging technologies and redesign workflows; governance and risk management that includes compliance consulting and talent support; value realisation to drive cost savings and business outcomes; and broader digital transformation services to set roadmaps and manage change.
Central to Genpact’s playbook is an ambition to industrialise AI development. In a corporate announcement on 28 January 2025, Genpact unveiled its AI Gigafactory — described as an enterprise‑scale accelerator intended to move pilots into production. The company characterised the Gigafactory as a “factory” model that combines industry expertise with proprietary agentic solutions, data marketplaces, engineering libraries and a responsible‑AI framework designed to manage the risks of goal‑oriented agents. The press release set an ambitious talent target — a pool of more than 25,000 AI builders — and highlighted partner collaborations, including Databricks, as ways to speed deployments.
The theory of the Gigafactory is mirrored in product form by Genpact’s Service‑as‑Agentic‑Solutions portfolio and, in particular, the Genpact AP Suite. Formally announced in June 2025, the AP Suite is promoted as a modular, agentic approach to accounts payable that pairs pre‑trained self‑learning agents with human expertise. According to the company, the suite’s four modules — AP Capture, AP Advance, AP Trace and AP Assist — are designed to automate invoice ingestion, resolve exceptions, manage supplier interactions and detect anomalies. Genpact claims outcomes of up to 80% touchless processing and up to 90% of supplier queries resolved automatically, plus benefits such as improved early‑discount capture and near‑zero duplicate payments.
Those are significant claims, and Genpact’s corporate materials and partner announcements repeatedly stress that these outcomes depend on orchestration: the coordinated work of AI agents, human specialists and platform integrations. The vendor’s expanded alliance with ServiceNow, announced in February 2024, is cited as an example of how embedding workflow and AI capabilities into Source‑to‑Pay processes can reduce cycle times — the announcement referenced a client example in which procurement cycle time fell by around half. Similarly, Genpact’s media briefings cite Microsoft Azure, Snowflake and other cloud and data partners as technology anchors that enable scale and governance.
But the wider evidence base urges caution. HFS Research and others continue to report low AI maturity across many organisations, particularly when it comes to scaling agentic or generative initiatives beyond pilots. Talent remains a bottleneck: Duggal acknowledged to the conference that “core talent is tough to find. They are limited in number. And they’re expensive. Once you get the talent, can you retain it? Can you make it exciting for them?” The answer, industry analysts argue, requires not only investment in people but also new operating models that distribute ownership — from BOTT (Build, Operate, Transform, Transfer) arrangements to virtual or assisted captive models and the emergent “agentic inter‑sourcing” approach that Genpact describes.
There are other practical and reputational considerations. Genpact frames the AI Gigafactory and agentic suites inside a responsible‑AI governance posture; the company’s public materials emphasise controls to manage agentic risk. Nonetheless, agentic systems — by design autonomous and goal‑seeking — raise novel questions over auditability, accountability and vendor lock‑in that GCCs and their parent organisations must answer. Independent verification of vendor claims is, for now, uneven: case studies and customer quotes in vendor materials are persuasive but do not substitute for broad‑based, third‑party evaluations of enterprise outcomes at scale.
For GCCs intent on becoming innovation engines, the path looks incremental and composite. Success is likely to come where three elements align: investment in high‑quality multidisciplinary talent; pragmatic governance and integration (both technical and contractual); and partnerships that combine domain knowledge with platform scale. Genpact’s propositions — its Gigafactory, the AP Suite and an expanding partner ecosystem — are explicitly designed to meet those needs. According to the company, they shorten time‑to‑value by standardising delivery, accelerating experimentation and providing pre‑built agentic capabilities.
Whether that formula can shift industry practice broadly will depend on how enterprises confront the hard work of organisational change as much as on the capabilities of vendors. GCCs, HFS and others agree on the direction of travel; the lingering question is when and at what scale the benefits of agentic AI will be realised across the majority of centres, rather than in a narrower set of early adopters. As the market moves from theory to deployment, the answer will reveal much about where value truly sits — in algorithms, in organisational design, or in the often messy interface between the two.
Source: Noah Wire Services
