SAS CEO: Responsible AI Deployment Key to Success

As enterprises race to deploy AI agents, the conversation is shifting from capability to control. Jim Goodnight, co-founder and CEO of US-based analytics firm SAS, argues that the real challenge is no longer building intelligent systems, but operating them responsibly at scale. In this interview with Sudhir Chowdhary, he explains why governance, transparency and engineering discipline will define the next phase of AI adoption – and why countries like India are well positioned to lead in managing AI-driven decision systems. Excerpt: We’ve always been in the business of helping organisations make better decisions with data, and that mission doesn’t change. What has changed is the user experience and the speed. In an agentic AI era, analytics can’t just produce insight. Analytics has to operate inside real business workflows with the right controls. Our role is to help customers move from experimentation to dependable deployment where outcomes are transparent, governed and repeatable, especially in high-stakes industries. When software takes action, the standard goes up. With autonomy, small errors can compound quickly, so you need engineering discipline, with governance, traceability, security, and clear accountability for what the system did and why. The goal shouldn’t be automation for its own sake. The goal should be better decisions delivered at speed that people can stand behind. Some are ready, many are not. That’s why governance matters. In regulated environments like banking , insurance, life sciences and government, trust isn’t optional. Before you delegate decisions to an AI agent, you need guardrails: where nondeterministic AI is appropriate, where it isn’t, how decisions are logged, tested and monitored, and when a human must be in the loop. There was a rush to generative AI when LLMs were introduced to the market, even before it was ready for production enterprise application. Fear of missing out drove a lot of spending that wasn’t tied to a clear ROI. Real ROI comes when AI is integrated into production decisioning securely, at scale, with performance you can measure and governance you can prove. We encourage customers to start with clear business problems, build the data foundation, and then deploy AI in places where the value is durable – places like risk reduction, faster cycle times and better service. Being privately held has allowed us to prioritise customers and employees as our most important stakeholders. It has let us invest for the long-term, especially in foundational areas like platform engineering, industry solutions, security and responsible innovation. We’ve built deep foundations in customer- and employee-focused culture that will carry forward through our next chapter of growth. Pursuing IPO readiness sharpens transparency, accountability and scalability across the business. It makes us a stronger company for customers, partners and employees. It helps us ensure that after 50 years of growth and profitability, we are built to support the next 50 with the same long-term mindset. India has the talent depth to do it, especially in data engineering, analytics and software development. I do think it can play an important role as AI becomes more operational. The challenge isn’t just building agents; it’s operating them responsibly at scale, with governance, reliability, security and cost discipline. The organisations that combine strong engineering with those enterprise standards will be the ones that succeed. We’ve built meaningful R&D and delivery capabilities in India over many years, and that’s important. It’s also a growing market where customers are moving quickly from AI experimentation to deployment. Our focus in India is the same as everywhere else: helping organisations apply AI to real industry problems with transparency and strong governance, so results are dependable.