This two-part series on boundaryless performance examines how AI is blurring the lines between work and learning, transforming the very nature of performance support. Part one, ‘From workflow learning to boundaryless performance’, examines these blurring boundaries in detail and introduces the ‘triage model’ adopted by leading organisations.
In part two, we examine real-world use cases of boundaryless performance, alongside unsolved challenges and future implications.
Real-world examples of boundaryless performance
From sales support to legal research, we examine how progressive organisations are implementing boundaryless performance.
Sales enablement: Continuous performance evolution
Sales organisations face a perpetual challenge: keeping field representatives updated on product changes while maintaining focus on customer relationships.
Leading organisations are implementing proactive notification systems with conversational exploration. When marketing releases new collateral, the AI assistant notifies the representative during their next scheduled check in, shares the document directly in their collaboration tool, and stands ready to answer questions as they review it.
This allows representatives to ask questions about changes in real-time without needing to search for documents or wait for responses. The result is faster adoption of new messaging and elimination of the artificial boundary between ‘learning about’ new information and ‘using’ it.
Cross-functional support: Dissolving organisational silos
Another challenge organisations face is helping employees access expertise across multiple domains without knowing the organisational structure. The triage model (explained in part one of this series) connects users to specialised knowledge bases through a single interface, allowing them to shift between topics seamlessly.
Here, the artificial divisions between organisational functions dissolve. Users simply ask questions and get answers without navigating departmental boundaries.
Legal research: The verification challenge
Not all implementations have been unqualified successes. A lawyer at a major firm recently shared her experience using AI to analyse complex case law. While the AI could process vast amounts of material quickly, she found herself ‘doing the work twice,’ using AI analysis but then verifying everything manually due to accuracy concerns.
What made this particularly challenging was that she needed expert knowledge to recognise errors; she couldn’t delegate verification to junior staff. This example shows both the promise and limitations of boundaryless performance in complex analytical tasks, where verification requires the same expertise the system was meant to supplement.
Truly boundaryless performance requires integration across multiple systems that often weren’t designed to work together.
Challenges of boundaryless performance
While the vision of boundaryless performance is compelling, significant challenges remain.
Content management challenges
Organisations struggle with document versioning and information currency. Constantly uploading new versions of policies without removing outdated ones increases the risk of AI hallucinations as versions get mixed up.
Additionally, large language models process lengthy documents differently, giving less weight to information that appears later. Strategic chunking of information becomes essential but requires new approaches to content management.
Expertise and verification challenges
The people best able to verify AI outputs are often the very experts the system was intended to augment, creating a paradox. This raises important questions about expertise development: How do junior employees develop deep knowledge if AI provides answers before they learn to solve problems? There’s justified skepticism about promises of rapid improvement, as progress is incremental rather than transformational.
Integration challenges
Truly boundaryless performance requires integration across multiple systems that often weren’t designed to work together. Legacy systems, data silos, and incompatible knowledge bases create artificial boundaries that limit the seamless experience users expect.
Quote: The question isn’t whether boundaryless performance will arrive, but how we shape it to enhance, rather than diminish, the human dimension of work.
Future implications
What does boundaryless performance mean for organisations and workers in the years ahead?
Organisational design implications
Despite advances in virtual collaboration and AI support, human interaction remains essential for certain types of knowledge transfer. The small social learnings that happen when a junior team member observes experienced colleagues aren’t easily replicated digitally. Organisations may need to balance remote flexibility with intentional in-person interactions that facilitate tacit knowledge transfer.
As boundaryless performance systems mature, the role of human experts will evolve to focus more on verification, edge cases, and innovation rather than routine information provision.
Skill development evolution
The skills needed for success in a boundaryless performance environment differ significantly from traditional models. Meta skills like effective system interaction, critical evaluation of AI outputs, and verification strategies become as important as domain knowledge itself.
Learning pathways must evolve from systematic knowledge acquisition to guided application with AI support.
Workplace dynamic shifts
When novices can access expert-level support through AI, traditional hierarchies based solely on knowledge possession flatten. Expertise becomes less about what you know and more about how you apply knowledge, interpret AI outputs, and handle novel situations.
Collaboration patterns are evolving, emphasising uniquely human capabilities that AI systems can’t replicate.
Is boundaryless performance a reality or aspirational?
Boundaryless performance isn’t a binary state but a spectrum. Organisations are at different points in the journey, and the most successful implementations focus on niche, specific domains rather than generalised systems.
In our professional experience, organisational value ultimately comes down to output. When tools integrate so deeply that they become part of the output process itself, the distinction between performance support and performance naturally dissolves.
The technical foundations for boundaryless performance exist today, but implementation complexity varies dramatically. The dissolution of boundaries occurs most readily where knowledge is structured, consequences are modest, and verification is straightforward. The human element remains essential not just for verification but for the social dimensions of learning and performance.
How should practitioners move forward with this? We recommend experimenting with boundaryless performance approaches in specific use cases while building the foundations for broader implementation.
The question isn’t whether boundaryless performance will arrive, but how we shape it to enhance, rather than diminish, the human dimension of work.
This article was co-authored by Teresa Rose, Director of Consulther Limited and Dr Markus Bernhardt, a globally recognized AI strategist and tech visionary known for his balanced approach to technology, innovation, and learning.
