Artificial Intelligence in IFSC - Survey 2026 Report

Artificial Intelligence in the GIFT IFSC Adoption, Maturity & Governance Survey Report July 2026 INTERNATIONAL FINANCIAL SERVICES CENTRES AUTHORITYContent Introduction and Executive Summary ........................................................ 3 Survey Design and Methodology .................................................................. 5 The State of AI Adoption ................................................................................. 6 Drivers, Barriers & Investment ....................................................................... 7 How AI Is Built .......................................................................................................8 Risk, Governance & Use.................................................................................. 10 Looking Ahead ................................................................................................... 12 Status in Major Sectors .................................................................................. 13 Conclusion .......................................................................................................... 16AI in IFSC | Survey Report 2026 Introduction and Executive Summary The International Financial Services Centres Authority (IFSCA) is a unified regulator for financial products, financial services and financial institutions in India’s International Financial Services Centres (IFSCs). Currently, GIFT IFSC is the maiden IFSC in India. As part of its mandate to develop a world-class financial centre while advancing innovation and safeguarding stability, IFSCA periodically assesses how emerging technologies are reshaping the regulated landscape.

This report presents the findings of the IFSC AI Survey 2026, which maps the adoption, maturity and governance of Artificial Intelligence across the regulated ecosystem.

Artificial Intelligence (AI) has rapidly shifted from a frontier technological exercise into a core computational layer reshaping global financial landscape. AI in the IFSC is in confident, accelerating transition. In a single year the respondent base doubled, Generative AI entered the institutional mainstream, and entities settled into a pragmatic, efficiency-first approach — while governance and assurance build in step.

Overall, the findings indicate that the GIFT IFSC ecosystem has moved from AI exploration in 2025 to early-stage operationalization in 2026, marking a significant increase in AI maturity while highlighting the need for governance, talent development, and regulatory guidance.

2025 → 2026 at a glance 2025 and 2026 Survey Results Comparison at Glance 82% Operational efficiency (top driver) 64% 74% Data privacy (top risk) 54% 57% Employees using public AI tools 52% 44% Third Party AI development 23% 35% Formal audit mechanism in place 10% 2026 2025 International Financial Services Centres Authority 3AI in IFSC | Survey Report 2026 Key Findings ▪ AI adoption is broad and advancing. Entities are exploring and piloting across all three AI technology classes – Predictive AI, Gen AI and Agentic AI. With 65% of entities already exploring/ implemented, Gen AI has achieved the deepest engagement whereas Agentic AI represents the next frontier, with early movers already piloting.

▪ Investment is gaining momentum. 31% of entities have invested in or are actively scaling AI spend, with a further 29% planning investment, a clear sign of growing institutional commitment. ▪ Operational excellence leads the agenda. Operational efficiency and process automation is the standout driver, cited by 82% of entities, well ahead of all other motives. AI in the IFSC is used for productivity gains across internal operations, risk and compliance.

▪ Employee AI use is widespread. 57% of entities cited that their employees are using AI tools, including widely used platforms. Entities are actively developing usage policies suited to their risk and compliance environments.

▪ Governance frameworks are actively developing. Entities are building AI accountability structures at pace, with a range of approaches tailored to organisational size and model.

Human-in-the-loop oversight is the most widely adopted production safeguard. ▪ Data stewardship is the top priority. Data privacy and protection is the highest rated concern, reflecting the IFSC ecosystem’s commitment to responsible and compliant AI deployment.

International Financial Services Centres Authority 4AI in IFSC | Survey Report 2026 Survey Design and Methodology A structured web questionnaire was designed, covering adoption, drivers, capabilities, risk, governance and outlook. The survey drew roughly double the participation compared to 2025.

Both multiple-choice and open-ended questions were included. All responses are anonymised and analysed only in aggregate. No entity-specific data is disclosed. Free-text examples are paraphrased or generalised to preserve confidentiality. Percentages are rounded, multi-select items sum to more than 100%.

Respondents by Sector Capital Market Intermediaries 27% Fund Management 20% Banking (IBUs) and PSP/ PSO 19% Insurance (IIO/IIIO) 18% Ancillary/TAS/BATF 15% Aircraft and Ship Leasing 10% Others 6% Market Infrastructure Inst. 3% Limitations • Voluntary participation and self-selection. Participation was voluntary. Entities already engaged with AI may have been likelier to respond, which can overstate ecosystem-wide adoption.

• Self-reported data. Responses reflect respondents' own assessments and were not independently audited or verified by IFSCA. • Uneven sample size. While the sample doubled year-on-year, sub-group counts remain small.

• Point-in-time snapshot. Findings on AI capabilities and entity practices represent the position at the time of the survey.

International Financial Services Centres Authority 5AI in IFSC | Survey Report 2026 The State of AI Adoption With responses spanning verticals - fund managers, capital-market intermediaries, banking units, insurers, finance, leasing and fintech, the ecosystem shows strong and growing engagement across all three AI technology classes. Generative AI demonstrates the deepest penetration, with close to two-thirds of entities at least exploring it and 17% already running it in production, ahead of Predictive AI and ML on the production metric. This reflects a notable acceleration from 2025, when AI engagement was characterised predominantly as early-stage and exploratory.

Agentic AI, AI systems capable of autonomous decision-making and action, represents the next wave of adoption. While 45% are not yet exploring it, a meaningful cohort has already moved to piloting or production, positioning the IFSC in forefront in emerging technology class.

AI Maturity Agentic AI 45% 35% 10% 7% 3% Generative AI 34% 29% 15% 17% 6% Predictive AI / ML 39% 32% 11% 15% 3% Not Exploring Exploring Piloting In Production Scaling By business function, AI activity is strongest in Risk & Compliance, covering AML-CFT, KYC and fraud detection, and Internal Operations including HR, legal support and IT. Customer-facing applications represent a strong and growing third category. With production deployments and maturity of workflow automation, the use cases are developed for delivering efficiency.

Deployment by Business Function Internal Operations Customer-Facing Risk & Compliance Credit & Lending Trading & Markets 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% In Production Piloting Exploring Not Exploring International Financial Services Centres Authority 6AI in IFSC | Survey Report 2026 Drivers, Barriers & Investment With responses spanning across verticals – fund managers, capital-market intermediaries, banking units, insurers, finance, leasing and fintech, the ecosystem shows strong and growing engagement across all three AI technology classes.

Operational efficiency is the standout driver and widened its lead since 2025 (64% → 82%). Cost reduction, customer experience, and regulatory compliance form a strong second tier. The main barriers are data quality and regulatory clarity, a lack of executive buy-in is the least-cited, implying that leadership intent is not the obstacle.

Drivers Barriers Regulatory uncertainty 41% Operational efficiency 82% Data availability & quality 41% Cost reduction 35% Lack of responsible-AI 34% tools Customer experience 33% AI/ML talent & expertise 27% Legacy infrastructure / 20% compute Regulatory compliance 28% Procurement & vendor 11% management Competitive advantage 26% Executive buy-in 6% Investment in AI across the IFSC is on a clear growth trajectory. A combined 60% of entities have already invested, are scaling, or are actively planning AI investment, with the remaining 40% continuing to assess the optimal approach for their business model and scale. Among entities that have committed budgets, the most common allocation is 1–5% of their IT spend.

Investment Status AI share in IT Budget 11% 9 18 Less than 1% 20% 40% Not planning 1% - 5% Planning 18 5% - 10% Invested More than 10% Scaling 29% 29 International Financial Services Centres Authority 7AI in IFSC | Survey Report 2026 How AI Is Built The IFSC ecosystem’s approach to building AI reflects the pragmatism of a diverse, global financial centre. Off-the-shelf SaaS tools and customised vendor solutions, including retrieval-augmented and fine-tuned applications, are the most commonly adopted approaches, allowing entities to move quickly with managed risk. Sourcing capability from parent organisations is also significant, enabling IFSC entities to leverage group-wide investments in AI infrastructure and models. In- house model development from scratch.

Methodology used for AI Build Model/ Platform used Vendor / off-the-shelf 44% OpenAI (GPT, Codex) 37% tools 23% Google (Gemini, PaLM) 26% 13% Group / collaboration Anthropic (Claude) 25% 16%

Provided by Parent 18% Organisation 11% In-house build 25% Microsoft Copilot 3% Meta (LLaMA) 2% 2026 2025 The clearest year-on-year shift is in delivery wherein entities have moved decisively toward vendor and off-the-shelf tools (23% → 44%) as managed offerings matured, while building in-house has narrowed. OpenAI leads foundation-model adoption at 37% of entities, with Google and Anthropic closely behind. The ecosystem draws on a healthy mix of providers, reflecting the diversity.

Microsoft Azure, Google Cloud and AWS anchor the cloud infrastructure picture where dedicated environments are in use. The plurality using parent-group or no dedicated infrastructure reflects the operational efficiency of leveraging established group-wide platforms.

Cloud Infrastructure used Microsoft Azure 19% Google Cloud (GCP) 18% No dedicated infrastructure 17% Amazon Web Services 17% Parent Organisation infra 10% On-premises 6% Private / Hybrid 3% International Financial Services Centres Authority 8AI in IFSC | Survey Report 2026 AI talent profiles across the IFSC are evolving in line with the maturity of the ecosystem. 44% of entities have dedicated AI technical staff in place or draw on specialist teams within the group.

On data governance, one-in-four entities have formal data-governance controls and privacy- compliance frameworks in place. Encouragingly, several entities reported mature governance approaches including strict data compartmentalisation and continuous validation, and ISO/IEC 42001 certification.

Dedicated AI Staff Data Governance Measures Data governance controls 27% 2% 7% No measure in place 27% 17% Governed by Parent 56% 27% Organisation 18% Strict compartmentalisation & 25% privacy checks Continuous quality 20% validation & monitoring None (0) Bias / representation Rely on Parent Organisation 12% testing 1 to 5 6 to 20 Data lineage tracking 10% More than 20 International Financial Services Centres Authority 9AI in IFSC | Survey Report 2026 Risk, Governance & Use Risks Data privacy and protection is the highest-rated technical consideration, cited by 74% of entities, a reflection of the IFSC ecosystem’s strong commitment to data stewardship and compliance. Model drift, data integrity and reliability of AI outputs form a closely clustered second group, areas where entities are actively putting in place validation and monitoring controls. Awareness of explainability and fairness considerations, while currently lower-ranked, is expected to grow as entities move to higher-impact, customer-facing deployments.

Technical Risks Systemic Risks Reputational risk (client- 56% 74% facing) Data privacy & security 54% AI-enabled market abuse 52% Vendor concentration 38% 21% Explainability / 2026 black-box 2025 41% Copyright / IP disputes 28% Financial exclusion 16% 16% Bias & fairness 23% Herd behaviour / flash 15% crashes At the system level, entities give greatest weight to reputational dimensions of AI deployment in client-facing contexts (56%) and to the importance of market integrity (52%). Vendor concentration and IP considerations feature prominently. The risk landscape reflects a sophisticated and alert ecosystem, with entities proactively identifying and managing the systemic dimensions of AI adoption.

Governance AI governance frameworks across the IFSC are developing at pace, with entities adopting accountability structures suited to their scale, business model and group structure. Where AI governance is formally assigned, it most commonly sits with a parent-group or a Chief Data/Technology Officer, reflecting the practical reality of how global financial groups organise AI responsibilities. A growing cohort of entities has established dedicated AI Governance Committees or CAIO roles, signalling an increasing commitment to structured and independent AI oversight.

International Financial Services Centres Authority 10AI in IFSC | Survey Report 2026 Governance Responsibility 35% Governed by Parent 20% Organisation 10% No formal responsibility yet 18% Formal audit in place Chief Data / Technology Officer 15% 2025 2026 Chief Risk / Compliance Head 8% 57% AI Governance Committee / 7% CAIO 52% Chief Executive / ExCo 6% Employees use public AI tools Decentralised (business units) 3% 2025 2026 Formal audit has tripled in share, from 10% in 2025 to 35% in 2026 whereas everyday use of public AI tools is widespread and increasingly managed (52% → 57%), shifting from informal use toward, monitored access.

AI in practice: use cases ▪ Document intelligence. Contract and agreement summarisation, intelligent document retrieval, automated statement-of-account processing, PII redaction, and data extraction with verifiable source referencing.

▪ Service automation. Internal AI helpdesks, enterprise chatbots, voice assistants, and AI- driven query and complaint handling. ▪ Productivity and engineering. Enterprise GenAI assistants for drafting, research and summarisation, AI coding assistants, and early agentic task-organisation tools deployed with human oversight.

▪ Risk, underwriting and analytics. Machine learning for creditworthiness assessment, persistency prediction and customer segmentation, fraud monitoring systems, guarantee and policy vetting with anomaly alerts, and AI-assisted underwriting decision support.

International Financial Services Centres Authority 11AI in IFSC | Survey Report 2026 Looking Ahead Entities take a measured, evidence-based view of AI’s workforce impact. The most widely held (32%) expectation is that AI will drive significant re-skilling and job transformation rather than large net reductions. A further 10% expect AI to generate net new roles through new product lines and capabilities.

Workforce Impact by 2030 Too early to estimate 48% Re-skilling & transformation 32% Net increase in roles 10% Net reduction in roles 6% Limited / no change 4% A wide base of Entities (48%) responded for regulatory clarity as leading priorities with 48% seeking clarity on applicability of existing rules on AI and 46% seeking Principles-based guidance, reflecting both the maturity of AI-deployment pipelines and the value entities place on clear, enabling guidance from IFSCA.

Guidance sought from IFSCA Clarify how existing rules apply to AI 48% Principles-based guidance (fairness, explainability) 46% Harmonisation across Indian & global regulators 33% Third-party-risk guidance for AI vendors 33% No additional intervention 25% AI regulatory sandbox 25% International Financial Services Centres Authority 12AI in IFSC | Survey Report 2026 Status in Major Sectors Banking Banking units show AI adoption concentrated in internal operations and risk/compliance, with limited movement into credit, lending or trading workflows. Maturity is uneven across AI types.

GenAI and Predictive AI show meaningful traction (around half of banks exploring or beyond) and Agentic AI shows momentum.

AI Adoption Maturity –Banking (IBUs) Deployment –Banking (IBUs) Agentic AI 55% 35% 5%5% Internal Ops 60% Risk & Compliance 45% GenAI 30% 30% 15% 15% 10% Trading & Markets 10% Predictive AI 40% 15% 15% 25% 5% Credit & Lending 20% Not Exploring Exploring Piloting In Production Scaling Customer Facing 30% Adoption is constrained less by talent (many institutions lean on parent-organisation resources and staff) and more by data availability/quality and a lack of responsible-AI tooling. Data privacy dominates technical-risk concerns.

Barriers –Banking (IBUs) Risks –Banking (IBUs) Ethical AI tools 45% Adversarial/cyber 35% Legacy infrastructure 20% Model opacity 20% Lack of AI talent 10% Model drift 25% Regulatory uncertainty 40% Hallucinations 35% Data availability 45% Data privacy 85% Fund Management Fund Management Entities (FMEs) remain in an early adoption phase. Across Predictive, GenAI and Agentic AI combined, more than half of responses fall in 'Not Exploring' or 'Exploring', with very few entities scaled or in full production for any AI type. Where AI is used, it concentrates on internal operations and risk/compliance.

AI Adoption Maturity -FMEs Deployment –FMEs 4% 4% Agentic AI 50% 38% 4% Internal Ops 58% Risk & Compliance 50% GenAI 38% 33% 13% 17% Trading & Markets 25% Predictive AI 42% 33% 17% Credit & Lending 21% 4% 4% Customer Facing 42% Not Exploring Exploring Piloting In Production Scaling International Financial Services Centres Authority 13AI in IFSC | Survey Report 2026 Unlike Banking, the leading barrier here is a lack of in-house AI/ML talent, ahead of regulatory uncertainty and a lack of responsible-AI tooling, pointing to capability gaps rather than data infrastructure as the primary constraint for fund managers. Data privacy again tops technical-risk concerns, the highest proportion of any sector after Banking Barriers –FMEs Risks –FMEs Ethical AI tools 38% Adversarial/cyber 33% Legacy infrastructure 29% Model opacity 29% Lack of AI talent 46% Model drift 29% Regulatory uncertainty 38% Hallucinations 25% Data availability 33% Data privacy 83% Insurance Insurance entities show an early-stage maturity profile similar to FME, with deployment again concentrated in internal operations and risk/compliance. Maturity is broadly similar across the three AI types, though GenAI shows the most advanced footprint with few insurers already have it in production or scaling, more than for Predictive or Agentic AI.

AI Adoption Maturity -Insurance (IIO/IIIO) Deployment –Insurance (IIO/IIIO) Agentic AI 50% 41% 5%5% Internal Ops 59% GenAI 36% 18% 18% 18% 9% Risk & Compliance 50% Predictive AI 41% 32% 14% 14% Trading & Markets 27% Credit & Lending 32% Not Exploring Exploring Piloting Customer Facing 36% In Production Scaling Regulatory uncertainty is the leading adoption barrier, ahead of a lack of responsible-AI tooling.

Technical-risk perception is more evenly spread than in other sectors with data privacy leading as top risk.

Barriers –Insurance (IIO/IIIO) Risks –Insurance (IIO/IIIO) Ethical AI tools 32% Adversarial/cyber 23% Legacy infrastructure 27% Model opacity 23% Lack of AI talent 23% Model drift 27% Regulatory uncertainty 41% Hallucinations 36% Data availability 27% Data privacy 59% International Financial Services Centres Authority 14AI in IFSC | Survey Report 2026 Capital Markets Intermediaries Capital Market Intermediaries (CMI) show the broadest functional footprint of any sector: Risk & Compliance, Internal Operations and Customer-Facing applications are all in active use, with Trading & Markets also meaningfully represented.

CMI is also the most advanced sector by maturity, where majority of the respondents have Generative AI in production, and over a third have moved Predictive AI or GenAI beyond the exploration stage, well ahead of Banking, FME or Insurance on this measure.

AI Adoption Maturity -CMI Deployment –CMI Agentic AI 45% 18% 18% 12% 6% Internal Ops 55% Risk & Compliance 58% GenAI 30% 21% 15% 24% 9% Trading & Markets 33% Predictive AI 36% 30% 9% 21% 3% Credit & Lending 18% Customer Facing 52% Not Exploring Exploring Piloting In Production Scaling Data availability and quality are the leading barriers, ahead of regulatory uncertainty, suggesting infrastructure and data-readiness gaps are more pressing here than policy clarity alone. Technical- risk concerns are dominated by data privacy, but CMI stands out for elevated concern around model drift, close to double the rate seen in Banking or Insurance.

Barriers –CMI Risks –CMI Adversarial/cyber 24% Ethical AI tools 30% Model opacity 21% Legacy infrastructure 9% Lack of AI talent 21% Model drift 49% Regulatory uncertainty 39% Hallucinations 36% Data availability 52% Data privacy 73% International Financial Services Centres Authority 15AI in IFSC | Survey Report 2026 Conclusion The IFSC moves through 2026 with broad, deepening AI adoption, Generative AI in the mainstream, and tangible gains in efficiency, compliance and client service. The survey captures

an ecosystem that has moved with real momentum: adoption has broadened across all AI technology classes, Generative AI has entered the institutional mainstream, and entities are delivering tangible results, particularly in operational efficiency, compliance and client service.

Governance frameworks are building in parallel with adoption, reflecting the IFSC’s culture of responsible innovation. Entities are making thoughtful choices about how to deploy AI, leveraging group expertise, proven platforms and structured human oversight, and are actively investing in the talent and data infrastructure that will sustain this growth.

The survey reflects a collaborative relationship between IFSC entities and IFSCA. Entities look to the Authority for principled, enabling guidance. IFSCA intends to continue developing regulatory frameworks that supports innovation, preserves market integrity and consumer trust, and positions the IFSC as a global benchmark in financial services.

Tracked year-on-year, this survey is IFSCA’s ongoing commitment to have an objective view of how AI adoption and governance are advancing across the ecosystem and equipping the Authority to calibrate its support as the technology and its applications continue to evolve.

International Financial Services Centres Authority 16INTERNATIONAL FINANCIAL SERVICES CENTRES AUTHORITY 2nd & 3rd Floor, PRAGYA Tower, Block 15, Zone 1, Road 1C, GIFT SEZ, GIFT City, Gandhinagar, Gujarat – 382 355 www.ifsca.gov.in