Generative AI alone is insufficient for advanced decision-making
Generative AI (GenAI) is a strong tool, but it can’t run the show alone. Jim Goodnight, CEO and cofounder of SAS, said it clearly: GenAI doesn’t cover the full depth of analytics, data science, or business intelligence needed in a real enterprise environment. This type of AI is good at generating text or code, but when the goal is reliable, accountable decision-making across systems and teams, GenAI hits limits fast.
Businesses operate with complexity, regulated markets, global supply chains, multiple lines of business, and non-stop decisions with financial, legal, and ethical stakes. These challenges demand more than a smart chatbot. They require purpose-built frameworks where AI models, business rules, and workflows work together in a structured and auditable way. That’s what SAS is focusing on through its enterprise AI capabilities.
The takeaway here for executives is straightforward: Don’t expect GenAI to replace your decision infrastructure. Augmenting your processes with GenAI can improve parts of your workload, but the actual backbone of enterprise decision-making remains a structured combination of humans, AI, and clear logic flows.
Purpose-built AI architectures are key for workflow execution
Satya Nadella, CEO and chairman of Microsoft, took this a step further. GenAI can help outline plans, but real business results come from execution, and that takes serious systems. In his words, we’ve made “enormous progress,” but there’s still work to do. A plan generated by AI has to live inside a workflow that understands context, rules, and the organization’s end goals.
This is where AI needs to shift from isolated tools to integrated systems. The AI models that support planning, task management, and even decision assignment are already evolving. Nadella pointed out that some AI functionalities now reduce workloads from eight hours to 30 minutes. That’s a game-changer for output and efficiency. Early GenAI use cases focused on surface tasks like replying to emails or writing snippets of code. But we’re already seeing that current models can take more responsibility, like assigning tasks, tracking progress, and nudging execution forward autonomously.
For senior executives, this changes the conversation around AI investment. The organizations that win with AI are designing workflows that match their strategy and integrating GenAI where it actually moves the needle. It’s about building workflows that deliver, at scale, with speed and integrity.
SAS Decision Builder on Microsoft Fabric
SAS has built something practical with support from Microsoft, it’s called Decision Builder. And it matters. In enterprise settings, where decisions have scale and consequence, you can’t afford fragmentation between data, models, and execution. Decision Builder pulls these elements into one framework that runs entirely in the cloud using Microsoft Fabric and Azure.
The big benefit here is location. With Microsoft Fabric’s OneLake, the data remains in place, accessible, and unified. You don’t have to waste time transporting information between environments. Instead, users can design, test, and execute decisions directly where the data lives. It’s connected to Azure AI Services, so businesses can integrate generative AI capabilities, like large language models, into decision flows without breaking structure or adding risk.
Satya Nadella made it clear: pushing SAS Decision Builder into the Microsoft Fabric environment puts decision-making right next to the data. That’s efficient. It also broadens access by using Azure Marketplace, making it easier to deploy this kind of AI-supported decisioning across markets and teams globally.
For executives looking to reduce latency in decision cycles and improve governance over AI models, this is meaningful. You’re not bolting AI onto decisions after the fact. You’re embedding intelligence upfront, using a system designed for deployment speed, security, and clarity. That translates into better, faster calls inside your business, and more consistency along the way.
SAS Viya Copilot democratizes AI-assisted analytics
SAS Viya Copilot solves an issue a lot of businesses don’t realize they have, access to AI tools inside technical environments can be uneven. GitHub Copilot transformed coding productivity for widely used languages like Python and R. But SAS users, who often work inside regulated sectors using more specialized tools, didn’t have a similar advantage. That changes with Viya Copilot.
This assistant, built with Microsoft Azure AI Foundry, brings automation, speed, and intelligence into the SAS platform itself. It’s conversational, which means it can understand prompts and return usable outputs. More importantly, it writes, explains, and refines SAS code, which removes bottlenecks in analytics pipelines. That means analysts and data scientists can focus on building better models, instead of manually correcting syntax or stitching workflows.
It also supports full model pipeline development. Executives should care about this because it directly impacts the time needed to bring a model from concept to deployment. Speed is essential for capturing opportunities in real time. And with human-in-the-loop features baked in, you maintain oversight and auditability while still increasing throughput.
For leaders aiming to scale AI maturity across the business, not just within central data teams, this tool helps level the playing field. It gives more team members access to real, practical AI capabilities without compromising quality, governance, or productivity.
Quantum computing is poised to transform AI
We’re at a point where classical simulation can only take us so far. Most current AI systems use approximations, they estimate reality well enough to make useful predictions, but they’re still guesses based on limited models. Quantum computing can change that. It allows us to simulate the real world at a level of accuracy that simply isn’t possible with classical systems.
Satya Nadella, Microsoft’s CEO, sees this as critical for pushing forward in fields like healthcare, biology, and computational chemistry. These industries operate on complex, tightly interrelated systems that require true-to-nature modeling, not just statistical assumptions. With quantum AI, businesses won’t have to rely purely on observed outcomes. Instead, they’ll be able to model systems with high fidelity, test interventions safely, and move faster with higher confidence.
This is where the research is headed. Quantum simulation offers the capability to solve computational problems that are currently intractable. That unlocks possibilities for drug discovery, material science, logistics, and economic modeling, all of which depend on understanding systems in full dimensional depth.
Senior leaders need to monitor this development closely. Quantum isn’t ready for day-to-day enterprise tasks yet, but the trajectory is clear. High-potential use cases are being tested now. If your business relies on complex modeling or operates in a high-variance environment, this is not something you’ll want to lag behind on. The returns will be significant for those who move early, build capability, and partner with the right platforms.
Key executive takeaways
- GenAI isn’t enterprise-ready on its own: Leaders should view generative AI as a productivity lever, not a decision engine, complex enterprise decisions still require structured frameworks and logic beyond LLM capabilities.
- Execution needs system-level AI integration: C-suite teams must prioritize building scalable AI workflows that go beyond ideation, enabling real-time task orchestration and outcome delivery across the business.
- Decision-making speed improves near the data: Embedding decision intelligence platforms like SAS Decision Builder directly into cloud environments cuts deployment times and improves governance, leaders should integrate decision tools where data lives.
- Technical AI must be accessible to non-experts: SAS Viya Copilot shows that democratising AI accelerates model development, executives should invest in tools that empower analysts and business users without sacrificing quality or oversight.
- Quantum AI will reshape high-stakes modeling: Leaders exploring innovation in healthcare, science, or supply-chain management should monitor quantum AI developments closely, it will expand what’s possible in real-world simulation and prediction.
