How We Do It
Key Issues within AI & Human Driven Work From Anywhere (WFA) organizations
ZeroSpace Advies focuses on jointly shaping AI & Human Driven Work From Anywhere (WFA) organizations where value creation comes from people who do AI-driven work and experience that they:
- use AI differently and treat it as a colleague, coach, and/or sparring partner rather than a “simple tool.”
- are valued and rewarded for their specific skills, competencies, capabilities, and experiences and the way they apply them.
Most organizations have adopted AI, but few have reshaped themselves to capture its value. The evidence is clear: enterprise impact does not come from more tools or more pilots. It comes from embedding AI into the workflows, decisions, and systems where work actually happens (Gutierrez, 2026).
In 2023, Mollick et al. invented a term that describes the weird ability of AI systems to do some work incredibly well and other work incredibly badly. Mollick et al. (2023) called this the ‘Jagged Frontier’ of AI. Even small amounts of ‘jaggedness’ can cause problems that prevent super-smart AI systems from automating a task.
AI-generated outputs were often not reliable in practice. It often led to ‘AI slop’ / ‘AI drab’, i.e., low-quality AI-generated work that has shortcomings and limitations.
This forced AI orchestrators to double-check every output, cross-reference every citation, and debug the logic — leading to an unexpected time-consuming ‘second job’. On the other hand, AI output can look so slick and polished that humans are no longer able to critically evaluate it.
AI’s real value is in augmenting, not replacing humans. Increasingly employees see AI not as a tool but as an employee — a colleague embedded in day-to-day workflows (EY, 2026).
Current research (Ranganathan & Maggie Ye, 2026) shows that in practice, AI tools do not appear to reduce workloads, but rather make them increasingly intensive.
Once the excitement of experimenting with different AI tools has faded, employees may find that their workload has quietly increased and feel overwhelmed by everything they suddenly have to do (EY, 2026). Such an increase in workload can, in turn, lead to cognitive fatigue, burnout, and impaired decision-making.
In other words, the (personal) productivity gains initially achieved may give way to lower-quality work, staff turnover, and other problems.
Furthermore, the emergence, development, and implementation of “AI agents” — i.e., systems that can autonomously plan, act, remember, and learn to achieve predefined results — is expected to continue. Today, gig platforms are available — for example https://rentahuman.ai/ — where AI agents can autonomously “hire a human” to perform a task.
Aligning skills with rapidly evolving AI-driven work through retraining, upskilling, training, and continuing education often leads to an identity crisis. In fact, instructing employees to reinvent their profession is a way of telling them that they need to fundamentally change who they are — and how they behave — as professionals (EY, 2026).
For example, the “prompt engineer” role rose and declined rapidly as users developed more sophisticated prompting skills and tools became better at prompting themselves. This illustrates how quickly the landscape can shift.
Communicate not just what AI will do, but what humans will do that’s more valuable. Currently AI systems can’t replace creativity, human connections, and empathy.
As organizational learning systems built for 12 to 18 month cycles cannot keep pace with new skills that evolve every 12 to 18 weeks, the need to develop and activate stealth learning programs is accelerated.
Stealth learning (Lekanne Deprez & Tissen, 2002) takes place when learning is built into the work process and is not seen as something isolated from that process. It is fully embedded and integrated into the work process, ensuring that people are constantly kept up to speed.
According to Durman (2025), in a Human In The Loop (HITL) system, human judgment is directly involved in decision-making. In a Human On The Loop (HOLT) system, an AI tool can operate autonomously, but there is someone who monitors the process and can intervene if necessary.
Disclaimer
During the process of reshaping and implementing AI & Human Driven Work From Anywhere (WFA) organizations, they often experience an initial decline in productivity. According to the so-called productivity paradox (Brynjolfsson, 2024), productivity within these types of organizations usually declines before it improves.
Get connected with Frank
© Zerospace Advies bv | Contact | design by xpressie