Learning and development teams have become increasingly good at delivering knowledge. Organizations can build courses, document procedures, create videos, distribute product information, and make learning available to employees wherever they work. The harder question is what happens when someone actually needs to use what they learned.
That challenge sits at the center of Grasply and the work of founder and CEO Edmundo Barrientos.
In this episode of the Training Impact Podcast, Jeff Walter talks with Barrientos about what Grasply calls the knowing-doing gap, the distance between understanding information and being prepared to apply it. Their discussion moves beyond the increasingly familiar use of artificial intelligence to generate training content and instead considers how AI can help organizations create something that has traditionally been much harder to scale: meaningful practice.
For L&D leaders, training managers, and operations teams, that distinction matters. A learner can successfully complete a course without being prepared for an unusual procedure six months later. A salesperson can understand an objection-handling methodology without being comfortable using it with a prospect. A manager can know company policy without having practiced the difficult employee conversation required to apply it.
Barrientos sees simulation as a bridge between those outcomes. The goal is not to replace courses, learning management systems, videos, manuals, or other forms of knowledge acquisition. It is to give learners somewhere to go after they know the information, but before they are expected to perform successfully in the real world.
Barrientos brings roughly two decades of experience in education and learning technology to that challenge. His journey began in Chile, where he founded a company focused initially on helping K-12 teachers and students understand academic subjects. As the organization grew and he became responsible for developing employees, his attention shifted toward a different learning problem.
Teaching someone a concept is one challenge. Helping that person apply it is another.
That distinction eventually became foundational to Grasply. Barrientos later moved to the United States, where the company is now primarily focused, bringing with him experience across both Latin American and U.S. learning technology environments.
His perspective on those markets also reinforces the importance of context in learning. Latin America may share significant linguistic similarities across countries, but individual markets differ in culture, regulation, administration, and business practices. The United States, by contrast, offers a much larger market and a more developed ecosystem of specialized learning technologies.
Despite those differences, Barrientos sees the underlying L&D problem as remarkably consistent. Organizations possess enormous amounts of knowledge, but transferring that knowledge is only part of developing performance.
Grasply is built around addressing what comes next.
One reason the knowing-doing gap persists is that people forget information they do not use.
During the discussion, Barrientos points to the forgetting curve and the importance of repetition. The practical challenge will be familiar to almost any training leader. Employees may regularly encounter the most common aspects of a process, reinforcing that knowledge through everyday work, while unusual exceptions occur only occasionally.
A procedure learned today might not be needed for another six months.
When that situation finally appears, the learner is expected to retrieve information that may not have been reinforced since the original training.
This is where course completion and performance readiness begin to diverge. An LMS can document that someone completed training. An assessment can demonstrate that the learner understood the material at a particular moment. Neither necessarily proves that the person can retrieve and apply the knowledge later under realistic conditions.
Barrientos connects this challenge to repetition. Learning becomes stronger as people repeatedly engage with information and behaviors, gradually strengthening the pathways associated with them. Watching someone ride a bicycle can teach the basic concept, but learning to ride ultimately requires getting on the bicycle.
The same principle applies to workplace capability. Salespeople need opportunities to handle objections. Managers need opportunities to navigate difficult conversations. Technical learners need opportunities to work through operational procedures.
The problem for organizations is that practice has historically been much harder to scale than content.
Role-play requires other people. Hands-on technical training can require equipment, instructors, facilities, travel, and scheduling. Those approaches can be extremely valuable, but the resources involved often limit how much practice organizations can realistically provide.
Grasply uses AI to change part of that equation.
Organizations already possess much of the knowledge required to define successful performance through SOPs, manuals, methodologies, presentations, product information, and other resources. Grasply can use that existing organizational knowledge as source material for simulations where learners begin applying what they know.
This represents a different use of generative AI in learning.
Instead of only asking AI to create another piece of instructional content, the organization can use its existing content to create an experience. An SOP can provide the logic for a procedural simulation. Product information can help shape a sales scenario. An objection-handling methodology can become something a salesperson practices rather than simply studies.
For training leaders, this creates a more complete learning progression. Foundational knowledge remains necessary, but simulation adds repetition and application between instruction and real-world performance.
Barrientos also introduces an important concept for understanding why simulations do not always need to reproduce reality perfectly: cognitive fidelity.
Imagine a worker learning how to arrange beverage cases efficiently on a pallet. One training approach would require an actual forklift, cases, pallets, floor space, and supervision. If the primary learning objective is understanding how the cases should be arranged, however, the learner can practice the underlying problem with a much simpler representation.
The physical environment is different, but the cognitive task can remain relevant.
This idea has significant implications for learning experience design because highly realistic simulations can be expensive. If the objective is to reproduce the decisions, sequences, and problem-solving involved in a task, organizations may not always need to reproduce the entire physical environment.
A digital simulation can therefore provide meaningful practice even when the learner is using a mouse and keyboard rather than touching the actual equipment.
The goal is to identify what the learner needs to think through and practice, then provide enough fidelity to make that experience useful.
For organizations responsible for geographically distributed learners, the concept becomes particularly interesting. The principles behind extended enterprise training often require organizations to provide consistent learning experiences to people who may not have equal access to instructors, facilities, or equipment. While the episode does not describe Grasply deployments specifically within extended enterprise programs, its approach to scalable practice addresses a challenge that becomes increasingly important as learning populations become more distributed.
Sales training is one of the clearest current applications for Grasply. A new salesperson can learn the organization’s products and objection-handling methodology, then enter a simulation where those concepts must actually be used.
The learner can practice responding to objections without waiting for the situation to arise with a real prospect. That creates opportunities for repetition while reducing the consequences of mistakes during early skill development.
The same model can support managers preparing for difficult employee conversations. Knowing the policy is important, but navigating the human interaction requires a different kind of practice.
Grasply is also moving beyond conversational simulations into operational learning. Barrientos describes simulations involving equipment, including work associated with medical devices. A digital representation can include the controls, indicators, and mechanisms necessary for learners to practice the sequence of operating a device.
The learner can initially receive guidance from an AI coach and later attempt the procedure independently. When something is done incorrectly, feedback can occur while the learner is still engaged with the task.
For L&D and operations leaders, this is where the knowing-doing gap becomes especially tangible. The learner is no longer simply demonstrating that the correct answer is known. The learner is beginning to demonstrate that the knowledge can be used.
The operational applications of simulation become particularly relevant when physical equipment is expensive, specialized, or difficult to make available for training.
Barrientos points to medical devices as an example. Traditional hands-on training may require learners to travel to equipment or specialists to travel to learners, adding cost and logistical complexity. More importantly, access to the physical device may represent one of the learner’s first meaningful opportunities to practice the procedure.
Simulation can create an intermediate stage.
Learners can become familiar with controls, sequences, and decisions before they work with the actual equipment. They can make mistakes in a learning environment, receive immediate feedback, repeat the procedure, and arrive at hands-on training with greater familiarity.
Barrientos is clear that digital simulation does not eliminate the need for physical practice. Instead, it can improve readiness for it. That distinction is important for organizations evaluating simulation technology because the objective is not to digitize every aspect of training. It is to determine where digital practice can make the eventual real-world experience more productive.
Grasply’s longer-term vision takes that progression further.
Barrientos describes the potential to move from two-dimensional simulations toward three-dimensional virtual environments when greater spatial or physical fidelity is valuable. From there, he sees another possibility in 3D printing.
Rather than reproducing an entire machine, AI could eventually help generate models of the specific components learners need to manipulate. Those components could be printed and combined with a digital or virtual simulation, adding tactile interaction without requiring access to the complete piece of equipment.
This remains a future direction rather than a current Grasply capability, but it reflects a broader perspective on learning technology. Digital learning does not necessarily have to eliminate physical interaction. The two can increasingly work together.
That idea also connects with the role tactile experiences can play in learning. Physical interaction, whether through models, manipulatives, handwriting, or other activities, can add another dimension to how learners engage with information. The Training Impact Podcast has explored this concept previously through LEGO Serious Play, where physical construction becomes part of the thinking and learning process.
For Barrientos, the future may involve combining these modalities more deliberately. Digital simulation provides scalability and repetition. Virtual environments can introduce greater spatial context. Physical components can add tactile experience. Actual equipment remains available when the learner reaches the stage where full physical practice is necessary.
For L&D leaders, perhaps the most important takeaway from Grasply is what it suggests about the next stage of artificial intelligence in training.
Generative AI has already made it easier to summarize information, generate assessments, create instructional materials, and help learners access organizational knowledge. Those applications primarily improve how content is created and consumed.
Grasply is applying AI to what happens afterward.
That shift changes the question from “How quickly can we create training?” to “How effectively can learners practice what the training teaches?”
It also increases the strategic value of existing organizational knowledge. SOPs, manuals, product documentation, sales methodologies, and other resources can potentially become more than instructional or reference materials. They can provide the foundation for experiences where learners use that information.
This reinforces an important principle for learning leaders. AI does not remove the need for strong training structure or accurate organizational knowledge. It can increase the value of those foundations by making them useful in new ways.
The companion case study, Grasply: Powerful AI Simulation That Closes the Critical Knowing-Doing Gap, takes a deeper look at this learning architecture, including training structure, learner types, operational challenges, and best practices aligned with the LatitudeLearning Training Program Roadmap. It provides a more structured examination of how knowledge acquisition, simulation, repetition, and feedback can work together to improve readiness.
Barrientos’s work with Grasply offers a useful reminder that training effectiveness ultimately depends on what learners can do with what they know.
Courses remain important. LMS platforms remain important. Manuals, videos, presentations, and other resources remain important. They establish the knowledge learners need before application can begin.
But completion is not the final outcome.
For training managers and enablement teams, simulation creates an opportunity to add structured practice before learners encounter situations where mistakes affect customers, employees, operations, or business performance. For operations leaders, it creates the possibility of preparing people more thoroughly before scarce equipment or specialist time is required.
As artificial intelligence continues reshaping learning technology, this may become one of its more meaningful applications. The opportunity is not simply to produce more content at greater speed, but to make the practice that follows that content easier to create and scale.
Grasply’s focus on the knowing-doing gap places that opportunity at the center of the learning experience. Knowledge establishes the foundation, but repeated application is what begins turning that knowledge into performance.
Want to go deeper?
🎧 To explore the full conversation, listen to the Training Impact Podcast episode featuring Edmundo Barrientos of Grasply.
📄 Download the companion case study: Grasply: Powerful AI Simulation That Closes the Critical Knowing-Doing Gap
🌐 Learn more about Grasply on their website: https://www.grasply.ai/
