AI adoption is real, but so is the change required - lessons from an ASUG Talks podcast with SAP CEO Christian Klein

Is AI going to make your job better or make it disappear?

For SAP CEO Christian Klein, the answer is not simple, but it is clear — AI is not just another technology wave. It is a force multiplier that will reshape how businesses operate and how people work within them. And this time, incremental change won’t cut it.

Klein spoke with ASUG CEO and Chief Community Champion Geoff Scott in a recent ASUG Talks interview, and covered the following -

• AI adoption is real, but most companies are underestimating the scale of change required.

• Business process redesign and lack of data context are the biggest hurdles to AI adoption.

• Enterprise software isn’t disappearing, but it will be fundamentally reshaped by AI agents.

• The future workforce will shift from operating systems to designing and optimizing outcomes.

• Data sovereignty and global volatility are reshaping how enterprises architect systems.

• Millions of professionals that make up SAP’s ecosystem must evolve alongside agentic AI.

• SAP is going “all in” on AI, focusing on outcome-driven use cases rather than tools.

The gap between AI hype and enterprise reality

At the highest levels of business, AI has already won. Boards and executives see it as a miraculous lever — faster decisions, lower costs, smarter operations. But inside organizations, the reality looks very different. Klein said, 

Obviously, a lot of people today love the hype of AI but then when the hype hits reality, oftentimes there are certain adoption blockers.

Those blockers aren’t what most expect. They’re not about model performance or computing power. They’re structural. 

What I’m hearing from every company is you need to rethink how your business will run in the future - be it coding of software, testing of software… you need to redefine the business process.

That means AI cannot simply be layered onto existing systems. It demands something far more disruptive — a redesign of how work itself gets done.

Klein argues that too many organizations are trying to, 

go into a brownfield system and plug in some AI agents. They start with the outcome first. 

That approach fails because it treats AI as an enhancement rather than a transformation. The companies seeing real value are starting somewhere else entirely. 

Why AI fails without business context

Even as large language models (LLMs) become more powerful, Klein believes context is a critical limitation holding enterprises back.

Some of these modules are super strong on unstructured content but at the end, what they are lacking is business context of the data.

That gap is more than technical—it’s existential for enterprise AI. Generic AI can summarize documents or generate code. But it cannot, on its own, understand the intricacies of a company’s procurement workflows, financial controls, or supply chain dependencies. Klein said, 

Not all of these LLMs should have access to your mission-critical data of your company.

This creates a paradox as the most valuable enterprise data is also the most restricted. Without it, AI lacks relevance. With it, governance becomes critical.

SAP’s answer is to embed AI directly into business processes, giving models structured access to workflows, data relationships, and permissions.

Why SAP is going 'all in' on AI

Klein is hardly alone among tech CEOs in declaring an AI-first strategy. But his rationale goes beyond signaling by creating a sense of urgency.

The technology is ready. It’s mature and we need to deliver the best agents for those businesses.

Inside SAP, that shift is sweeping. 

All our product managers are now asked, ‘Hey, redefine how businesses would run in the future.

It’s not just product development that’s changing. SAP is transforming in software engineering, too. 

We are using the latest code generation tools because we also need to become more productive. How can you be credible to your customers if we are not also transforming the company inside SAP?

But perhaps the most significant shift is commercial.

We want to sell even more in the future outcomes, with AI solving business problems. 

That marks a departure from traditional enterprise software sales, which focused on features and modules. In an AI-driven world, value is measured not by what software does, but by what outcomes it delivers.

The rise of the autonomous enterprise

To understand how deeply AI could reshape enterprise work, consider the activity of cash collection, a complex process involving multiple systems, decisions, and human interventions. But in an AI-driven future, much of it becomes autonomous, taking over complex business processes. This is where the concept of 'agentic AI' becomes real.

Instead of software being used by humans, AI agents execute workflows themselves, guided by rules, data, and objectives.

But this shift introduces new challenges as well. Klein said, 

You need an authorization profile for an agent. You don’t want to have your financial data flying around in the whole company.

It also introduces new scale as organizations may soon manage not just human workers, but thousands of AI agents, each performing specialized tasks. The result is a fundamentally different enterprise architecture built on applicationsandcoordinated intelligence.

Is software dead? Not even close

Amid the rise of AI, a new narrative has emerged that traditional enterprise software is becoming obsolete. Klein rejected that idea outright. AI depends on ERP and line of business solutions.

Every agent needs a piece of software or business process context. Every agent needs data, and you need to store this data somewhere.

In fact, software may become even more important, not less. Without enterprise apps and data, AI won’t have anything to work on. But what will change is how software is built and used.

Klein predicted that agents will take over coding a lot of the features found in software. That doesn’t eliminate software — it transforms it from a user interface into an underlying system of record and logic. 

The software itself will not go away but it will look very different.

The hidden cost of global volatility

While AI dominates headlines, another force is quietly reshaping enterprise technology — geopolitical instability. The consequences are profound with a direct impact on data.

Customers want to better understand where their data is getting stored and who has access. That concern is driving architectural complexity. SAP’s response is to offer flexibility by allowing customers to choose where and how their systems run.

But the broader trend is clear — the era of simple, global IT architectures is ending—and AI will only intensify that tension.

The future of the SAP professional

For the millions of professionals working in the SAP ecosystem, the stakes are deeply personal. Klein acknowledged both the scale and the importance of that community.

We have over 9 million consultants all over the world. Without them, SAP wouldn’t run.

But their roles are about to change. In the past, much of SAP work involved customizing systems, writing code, and maintaining complex environments. That model is fading.

We need the ecosystem to go with us again. We want to build a platform where our customers and partners can build their own agents.

This opens the door to a different kind of work. Not coding in the traditional sense, but designing processes, defining outcomes, and orchestrating automation.

It’s not only about IT — it’s also about the business user.

Low-code and no-code tools will allow non-technical users to build and extend AI-driven workflows. That democratization could be one of AI’s most transformative effects.

Staying grounded in a world of change

For Klein, who started at SAP as a teenage intern, the pace of change is both familiar and unprecedented. Looking back, his advice is surprisingly simple.

Stay curious. Don’t believe that you know everything already.

That mindset may be the most important skill in the AI era because if there’s one certainty, it’s that the rules are still being written. And while AI is not coming for your job in the way many fear, it is coming for your tasks, your workflows, and the way your organization operates.

For SAP and its vast ecosystem, the path forward is clear — embrace the shift, redesign the business, and build for outcomes, not tools.