New Delhi, May 19 -- SaaS major Salesforce announced two India-focused initiatives aimed at strengthening its enterprise AI and cloud infrastructure strategy, as the company sharpens its focus on agentic AI adoption and localised enterprise capabilities.
At its Agentforce World Tour Mumbai 2026 event, the company unveiled Hindi language support for Agentforce Voice and announced the upcoming availability of MuleSoft on Hyperforce in India-moves designed to help businesses deploy AI-led customer experiences and manage enterprise integrations on locally hosted infrastructure.
The announcements come as enterprises increasingly seek AI systems capable of operating at scale while meeting data localisation and governance requirements.
Salesforce said Agentforce Voice in Hindi will enable organisations to deliver AI-powered multilingual customer experiences, supporting natural Hindi and Hinglish interactions. The capability is intended to improve accessibility and broaden AI adoption beyond metro markets into Tier-2 and Tier-3 regions, where voice-based interactions often serve as a primary digital interface.
Unlike conventional IVR systems dependent on predefined decision trees, Agentforce Voice uses conversational AI with action-oriented reasoning to support more contextual and autonomous interactions. The platform combines enterprise data, workflows and AI capabilities while operating within policy-driven guardrails designed for secure and consistent engagement.
The company also said it plans to introduce more Indian subcontinent languages in the coming months, building on support for over 25 languages already introduced globally for Agentforce Voice.
Dr. Satya Ramaswamy, Chief Digital and Technology Officer at Air India, said voice technologies are becoming increasingly important in multilingual markets such as India.
"Voice is emerging as a critical interface for the next generation of customer experience, especially in a multilingual market like India," he said, adding that local-language AI interactions could improve accessibility and personalized support at scale.
Alongside the voice announcement, Salesforce said MuleSoft on Hyperforce will become available in India, allowing enterprises to deploy APIs, integrations and AI workflows on local infrastructure.
The launch expands Salesforce's Hyperforce footprint in the country and is expected to appeal particularly to sectors with stringent regulatory and data residency requirements, including financial services, healthcare and the public sector.
With both control and runtime environments hosted within India, enterprises will be able to store and process integration data locally while accessing AI-powered workflow capabilities. The roadmap also includes planned capabilities such as MuleSoft Agent Fabric, Omni Gateway and Vibes, designed to help enterprises govern AI agents across systems and vendors.
Arundhati Bhattacharya said the growing use of AI across enterprises is increasing the importance of trusted integrations and governance frameworks.
"As AI agents multiply across the enterprise, the ability to govern and orchestrate them across systems and workflows will become foundational," she said.
The announcements signal Salesforce's broader India strategy as enterprises increasingly move from AI experimentation toward large-scale deployment and operational integration.
Salesforce's India initiatives also come amid a broader acceleration of its global AI strategy. Marc Benioff recently said the company expects to spend nearly $300 million on Anthropic tokens in 2026, underscoring the growing role of AI models in core business operations.
Speaking on the All-In podcast, Benioff said AI coding agents and Anthropic models are already improving efficiency across Salesforce's engineering organisation of around 15,000 developers. He added that AI tools are reducing development costs and accelerating product creation, while hinting at upcoming AI-led coding capabilities within Slack.
Benioff also said enterprises will increasingly require orchestration layers capable of intelligently routing tasks between expensive frontier AI models and smaller, cost-efficient systems as adoption scales.
Published by HT Digital Content Services with permission from TechCircle.