Salesforce unveiled Koa and expanded Agentforce at Dreamforce 2026 while Air India deepened its AI automation use. Here is what the announcements mean for enterprise AI.
Salesforce used Dreamforce 2026 to make a much bigger claim for enterprise AI: the next stage is not simply answering questions, but reasoning across company data and taking actions inside real business workflows.
One of the headline announcements was Koa, Salesforce’s first AI reasoning model. The company says it is built on Nvidia’s Nemotron technology and designed for enterprise tasks across sales, marketing and customer service. The goal is to make AI more useful inside the structured, permission-heavy environment of corporate software.
Salesforce also expanded its Agentforce strategy, which focuses on AI agents that can plan, reason and complete multi-step work. That approach matters because businesses are increasingly asking whether AI can move beyond chat interfaces and actually reduce the time employees spend on repetitive processes.
Air India is one of the most visible examples. Salesforce says the airline is expanding Agentforce across customer-service workflows. In earlier deployments, the company reported major reductions in refund turnaround times and faster passenger-name updates, showing why airlines are interested in automating complex service requests.
The opportunity is large, but so are the challenges. Enterprise agents need reliable access controls, accurate data, auditability and clear limits on what they are allowed to do. A helpful consumer chatbot can tolerate a mistake more easily than an AI system changing a booking, approving a refund or acting on a customer record.
Dreamforce 2026 therefore highlights a broader shift in business AI. The competition is moving from who has the smartest standalone model to who can connect models safely to real company systems. Salesforce wants Koa and Agentforce to become that operating layer, and Air India offers an early test of whether the idea can deliver at large scale.
Why this matters
This development matters because it changes the choices available to users, businesses or policymakers in a fast-moving area. The key next step is to watch execution, adoption and any official follow-up.
Related coverage
- Wipro Ai Capacity 20000 Workers What Happens Jobs
- Meta Muse Ai Agent Email Shopping Booking Privacy
- Openai Chatgpt Financial Services Launch
Source: The Indian Express
