Cornerstone Workforce AI: What It Is, What It Does, and Why You Should Start Now

Holger Mueller sat down with Cornerstone OnDemand's Suchi Upadhyayula in New York, on the day they launched Cornerstone Workforce AI. Here is what customers need to know.

What Cornerstone Workforce AI Actually Delivers

Let's start with what is new for existing customers. If you are on Cornerstone's learning platform today, meaningful AI innovation is arriving for you right now.

1. First, there is a course assistant. This is built directly into the learning experience and helps learners go deeper on topics they care about, complete quizzes, and engage with content in a more interactive way than was previously possible. It is a better learning experience, not just a chatbot bolted on top.
2. Second, there is an adaptive learning agent with role-play capabilities. This is significant. The ability to practice what you have just learned in a simulated scenario within the same platform closes a gap that traditional learning management systems have never been able to address. You learn. You practice. You retain.
3. Third, for customers on the talent platform, there are new assistive features for performance reviews. Nobody loves writing performance reviews. Managers dealing with 360 feedback across large teams lose meaningful hours every cycle. Cornerstone is now helping surface that work faster with AI assistance, both for the person writing the review and for the manager synthesizing feedback.

The Intelligence Plus Package

These capabilities land differently depending on your current setup. If you are a Learn Plus customer, you get meaningful new functionality out of the box at no additional cost. If you are looking for more agentic, outcome-based capabilities, that is where Intelligence Plus comes in. The tier structure is logical, and the on-ramp is designed to let customers build from where they are rather than rip and replace.

Custom Workflows and Forward Deployed Engineers

Beyond the packaged capabilities, Cornerstone is now offering something more bespoke. If you have a specific outcome in mind, say, a custom onboarding flow tailored to your organization's brand and business processes, Cornerstone will send forward-deployed engineers to build it with you. This is a meaningful shift. It is not a vendor handing you a product and leaving. It is a co-creation model that takes your nuances seriously.

This approach is gaining momentum across enterprise software, and Cornerstone is leaning into it correctly.

MCP Support and the Open Platform Play

One of the more technically interesting announcements is MCP support. Cornerstone has built a trusted data platform as its foundation, layered strong APIs on top of it, and is now exposing those APIs as MCPs. In practice, this means that if your organization is already running Claude, OpenAI, or another agent platform of your choice, you can connect directly to Cornerstone's data sources and capabilities via standard MCP protocols.

The result is a flexible, composable architecture. Your enterprise decides which agent platform to use. Cornerstone's data and functionality connect to it. The learning content meets employees where they already are, in Teams, in Slack, in whatever tools they use every day. That is the right design principle, and it is good to see it in practice.

Skills Without the Spreadsheet

One of the sharpest observations in the conversation was about how skills are actually built and detected. Nobody wakes up in the morning thinking about which skills they need. They wake up, go to work, and do tasks. A Java developer who needs to learn Python is not thinking about skill acquisition. They are thinking about getting their work done.

Cornerstone's approach to skills detection is task-based rather than declarative. Instead of asking employees to fill out yet another skills inventory spreadsheet, the system infers skills from what people are actually doing. That is the right direction, and it sidesteps one of the oldest failure modes in HR technology.

Making Your Content AI-Ready

Most Cornerstone customers have rich learning catalogs assembled over the years. The challenge is that a catalog sitting in a system is not the same as that catalog being accessible to AI. Cornerstone is launching assistive tools to help customers curate, select, and vectorize their existing content to make it AI-ready. User-generated content can be promoted to AI visibility with a single step. It is a practical, low-friction path to getting value from what already exists.

Start Now

The most important piece of advice from this conversation is simple: start the journey now.

There are still many organizations sitting on the sidelines, waiting to understand exactly how the AI inference works, whether compliance requirements are met, and whether the timing is right. Cornerstone's message is that they are here to help navigate those internal approval processes. But the deeper point is this: the innovation cycle in this space is no longer yearly. It is quarterly at best and monthly at worst. If you wait for a perfect moment, you will spend the rest of the year catching up to organizations that started six months ago.

You do not need to have it all figured out. You need to get started, learn from it, and build from there. That is exactly what Cornerstone is building toward with its co-creation model, and it is the right instinct for where enterprise AI is right now.

Human to the power of AI. That is the frame Suchi closed with, and it is a good one.