Have you ever written a piece of code, run it, and then asked yourself: "What if this software didn't just suggest things to me, but actually did them itself?" That is no longer a question for daydreaming developers. That is the problem SAP has just announced it has solved.
At SAP TechEd 2026 in Berlin, SAP introduced far more than a few new AI features. They announced a paradigm shift: moving from AI as an "apprentice" within the enterprise to AI directly operating the enterprise. At the center of this shift are two names: Joule Work and Joule Studio.
Join me, SilverZ, as we dissect what SAP is truly putting into developers' hands and why this could be one of the most pivotal turning points for enterprise AI in 2026.
From "Suggested AI" to "Self-Executing AI": SAP is Changing the Rules of the Game
For years, AI in enterprise software primarily served as an assistant: suggesting answers, summarizing data, and drafting text. Useful, but still requiring humans to click buttons, approve, and execute. Joule Work changes that endpoint.
As SAP describes it, Joule Work is where users input their intent and receive completed work through a seamless experience spanning the entire SAP ecosystem and beyond. In other words, you no longer "ask AI and then do it yourself." You state what you want, and the AI plans, coordinates, and executes the necessary steps to turn that intent into results.
To achieve this, Joule Work is equipped with agentic reasoning capabilities: it can plan and orchestrate the steps required by assistants and agents to complete complex, multi-step business processes across SAP systems. This is no longer linear automation. This is goal-driven autonomy.
Joule Studio: Empowering Non-Coders to Build Agents
The second notable highlight is Joule Studio. SAP is not just empowering professional developers. They are opening the door to business users—those who understand processes but cannot write code.
Joule Studio provides both no-code and pro-code tools to build and extend agents. Users can choose from over 70 different AI models rather than being locked into a single model. For customers in the early adopter program, they can even design and run n8n workflows directly inside Joule Studio.
This is a strategic shift. It acknowledges a truth that many enterprises have known for a long time: the person who understands the problem best is sometimes not the best coder. By giving them the tools to build their own agents, SAP is distributing AI capabilities closer to where the actual work happens.
SAP Business AI Platform: 50 Years of Context That AI Cannot Generate on Its Own
What makes an SAP agent different from a generic agent? The answer lies in context.
SAP built the SAP Business AI Platform as a single destination to build, contextualize, and govern AI across both SAP and third-party environments. This platform is anchored to the SAP Business Data Cloud and SAP Knowledge Graph, providing agents with reliable data and business context for reasoning.
The core lies in over 50 years of industry processes and expertise encoded into every agent action. An agent does not just know "what needs to be done." It knows whether doing so makes sense in this specific enterprise context, whether it complies with regulations, and where to halt if anomalies occur.
And there is a commitment every enterprise should remember: customer data remains customer data. SAP states that it does not share customer data with anyone to train or improve large language models. In an era where data is an asset, this is a weighty promise.
Governance and Interoperability: The Hardest Challenge of Agentic AI
An autonomous agent system without governance is not a solution. It is a risk. SAP seems to understand this very well.
Joule Work supports open protocols like Agent2Agent (A2A) and Model Context Protocol (MCP), allowing agents from different vendors to collaborate within the same business process. More importantly, agent invocations are handled through a governed agent gateway with enforced access controls.
At a higher level, the SAP Enterprise AI Gateway extends governance to all AI interactions across the enterprise: centralized authentication, policy enforcement, routing, monitoring, usage tracking, and auditing. This is the infrastructure any organization looking to scale agentic AI safely needs to have.
In other words, SAP is not just selling you tools for AI to take action. They are selling you systems to control AI when it takes action.
The International Space Station and Tabular Models: Symbols of the "Possible"
In its announcement, SAP mentioned an apparently peripheral detail that carries powerful symbolism: The International Space Station (ISS) is using SAP's tabular AI model.
This is the SAP RPT-2 model, the next generation of SAP's foundational tabular model lineup. Its deployment on the ISS—an extremely harsh environment where data and decisions can impact human safety—demonstrates that SAP wants to prove one thing: their AI doesn't just run in boardrooms. It can operate where reliability is paramount.
This is also a reminder that tabular AI—processing tabular data, the backbone of all ERP systems—is not a secondary concern. It is the foundation for AI to understand the true language of business: rows, columns, metrics, and the relationships between them.
Implications for Developers and Vietnamese Enterprises
You might be thinking: "SAP is a system for large corporations. What does this have to do with me?"
The answer is: a lot, in two directions.
First, this is a signal about skills needed in the near future. When SAP brings no-code and pro-code together into the same studio, and when agents can be built by business-savvy users rather than just programmers, the boundary between "developer" and "business user" is blurring. Vietnamese developers don't just need to know how to write code. They need to know how to design agents, understand business context, and practice responsible AI governance.
Second, this is an opportunity for Vietnamese enterprises to rethink automation. Many domestic companies are still struggling with step-by-step automation. SAP is demonstrating a different model: starting from intent, letting AI plan and execute, while humans retain the roles of goal-setting and supervision. That is the mindset required to compete in the digital economy.
Third, SAP's data policy is a reminder. As many Vietnamese enterprises remain hesitant about moving data to foreign AI platforms, the commitment not to use customer data for model training is a standard that domestic providers should also strive for.
When Intent Becomes Action
I have watched enterprise AI go through many phases: from scripted chatbots to suggestive assistants, and on to large language models. Each phase promised more than the last. But SAP TechEd 2026 marks a distinct difference: this time, the story is no longer about "what AI can say," but "what AI can do."
Joule Work turns intent into action. Joule Studio empowers more people to build. The SAP Business AI Platform provides context to make those actions meaningful. And the Enterprise AI Gateway ensures those actions remain within controlled boundaries.
Yet, like every technological revolution, success does not lie in the tools. It lies in those who know how to use the tools to solve real problems. And the biggest question remains: is your enterprise ready to grant AI the power to take action, or do you still just want it to make suggestions?
If tomorrow you had an agent capable of independently planning and executing a complex business process—from ordering and invoice processing to reconciliation—what task would you assign to it first, and what boundaries would you set to feel safe? Share your thoughts below, because in the era of agentic AI, those who know how to set the right boundaries will extract the most value!
0 Comment
Add your comment to this article