Background
As AI tools multiply across the enterprise, one of the biggest challenges teams face is not access, it is confidence. Knowing a tool exists and knowing how to use it effectively are two very different things, and the gap between them is where adoption stalls.
The GenAI Experience Lab (GenAI XL) is SAP CPIT's internal AI experimentation and adoption platform, built to help employees across the organisation explore, learn, and work with generative AI in a safe and governed environment. As Product Owner for the platform, running targeted upskilling sessions is a core part of my role , connecting the product to real usage by real teams.
The challenge
SAP's AI tooling landscape is broad and expanding. For teams outside of tech, the proliferation of tools creates a specific kind of friction: decision paralysis. Which tool do I use? For what? Is this the right one? That uncertainty often leads to underuse or inconsistent use, even when the underlying tools are genuinely useful.
People & Culture teams, in particular, work with sensitive information daily, which makes questions of data safety especially relevant when introducing AI tooling. The session needed to address both practical capability and responsible use in the same breath.
Session design
Rather than a passive demo, the session was designed as a guided walkthrough of the GenAI XL platform with participants following along live. The goal was to create genuine moments of "I could use this" rather than "interesting, but I'll figure it out later."
Five capability areas were covered in sequence, each connected to scenarios relevant to the team's actual work:
Effective prompting
Small changes in how a question is framed can significantly transform the quality and relevance of an AI response. The session explored the mechanics of this in practice, rather than as an abstract principle, using examples grounded in HR and people ops contexts.
Conversation sharing
One of the most underused features in team AI adoption: the ability to share conversations so colleagues can build on each other's work. This reduces duplicated effort and accelerates learning across the team organically.
Spaces for team collaboration
GenAI XL Spaces allow teams to work together with shared context, custom instructions, and a common knowledge base. For P&C teams coordinating on policy, onboarding content, or employee programmes, this creates a meaningful upgrade over one-off individual interactions with AI.
Web search grounding
Grounding AI responses in real-time information helps teams avoid outdated outputs and increases confidence in the results. This is particularly relevant for roles that need to stay current on policy changes, labour regulations, or industry benchmarks.
Data masking & responsible use
GenAI XL automatically masks PII in prompts before they reach the AI model, a feature that directly addresses the hesitation many P&C professionals feel about entering sensitive information into AI tools. Surfacing this early in the session helped shift the conversation from "should we use this?" to "how do we use this well?"
Navigating the AI landscape
A dedicated segment was spent helping participants develop a mental model for navigating SAP's AI tool landscape. Rather than presenting a rigid decision tree, the focus was on building intuition: understanding the distinct purpose of each tool so that the right choice becomes instinctive over time. For teams at the start of their AI adoption journey, this kind of orientation is often more valuable than any individual feature walkthrough.
Outcome
The measure of a good upskilling session is not whether participants walk away impressed -- it is whether they walk away ready to act. The questions that came up during the session reflected genuine engagement: people connecting capabilities to specific situations in their own work, not just acknowledging that the tool exists.
That shift from curious to capable is what makes these sessions worth running, and what GenAI XL is ultimately built to enable.
The questions they ask, the ideas they start connecting -- that is where the real value is.