Bridges Transitions for AI Role Changes

COMPEL Specialization — AITE-WCT: AI Workforce Transformation Expert Article 21 of 35

William Bridges’s Managing Transitions, first published in 1991 and now in its 4th edition (2017), introduced a distinction that the rest of the change-management field had missed. Change, Bridges argued, is the external event — the announcement, the restructuring, the new system, the role redesign. Transition is the internal, psychological process that the people experiencing the change must move through. The two are different. Organisations that treat them as the same — announcing the change and assuming the transition will follow — produce a characteristic failure mode: a population that looks compliant but has not actually let go of the old way, has not worked through the uncomfortable middle, and has not committed to the new beginning. The behaviour change the organisation expected does not appear, or appears thinly and decays quickly.

For AI workforce transformation, Bridges is indispensable in the role-redesign and redundancy workstreams (Articles 24–29) because those workstreams are almost always loss-laden. A role that existed before — with its skills, its identity, its psychological contract — ends. Something new begins. Between the two, there is a psychologically uncomfortable period the organisation must learn to support rather than rush past.

The three phases

The Ending. Before anything new can begin, the old must end. Ending is the work of letting go — of an identity, a routine, a set of skills that defined competence, a set of relationships that the old role supported. The ending work is not automatic. An employee whose role is being redesigned may understand the change cognitively and be unable to let go emotionally. They continue to practise the old role alongside the new, undermining the new while feeling loyal to their past self.

The Neutral Zone. The period between the ending and the new beginning. Bridges described it as a “no-man’s land” where the old no longer works and the new does not yet. It is characterised by ambiguity, low productivity, high anxiety, and sometimes high creativity. People in the neutral zone are learning but not yet performing; they are uncomfortable; they often appear “resistant” when they are in fact in transition.

The New Beginning. The moment of commitment to the new — not just the acceptance that the old is gone, but the active engagement with what the new requires. Bridges argued that the new beginning needs energy, and that organisations commonly underinvest in it, assuming that once the neutral zone passes, commitment will follow automatically.

The three phases are not neatly sequential for every individual. A person may be in the neutral zone on one aspect of their work and still in ending on another. The framework is a diagnostic and an intervention guide, not a timeline.

Why Bridges matters for AI role changes

AI role change has specific features that make Bridges particularly apposite.

• Identity content is high. A senior analyst who becomes an “AI-augmented advisor” has not just learned a new tool. They have renegotiated what their professional identity is. The question “what am I good at now” is central.
• The prior skill was hard-earned. Most redesigned roles involve displacing skills the incumbent took years to build. Letting go of those skills — even when the new role uses them differently rather than abandoning them — is a slow process.
• The uncertainty persists. Unlike a role change where the new role is fully defined on day one, AI-role change often has an evolving specification as the AI system matures. The neutral zone can last longer than in traditional transitions.
• The community is transitioning together. Entire professional communities — underwriters, radiologists, junior lawyers, customer-service agents — go through AI transitions simultaneously. The shared experience is both a support (peers understand) and an amplifier (collective anxiety).

An expert who applies a pure ADKAR or Kotter frame to a population going through these transitions will miss the psychological work. The learners will score well on Knowledge and Ability assessments and will feel, a year later, that nothing has really changed about how they see themselves. That gap is the Bridges gap.

Designing for the Ending

The Ending phase is the most under-designed. The organisational instinct is to minimise the ending — to skip quickly past what is being left behind and emphasise what is being gained. The instinct is wrong.

Three design elements support the Ending:

• Name what is ending, specifically. Not “your role is being transformed” but “the drafting work you do by hand today will largely be done by AI; your role will shift to review and client dialogue.” The specificity gives the person something to mourn, which is what lets the mourning complete.
• Honour what was. Acknowledge that the skills being displaced were valuable, that the way of working was valuable, that the contribution was valued. This is not ceremonial; it is practical. A person whose work is not honoured at its ending will carry resentment into the new role.
• Mark the ending visibly. A team ritual, a communication that closes the prior chapter, a structured farewell to the old process. Organisations feel awkward about this. Bridges argued that the awkwardness is the point — the ritual creates the psychological punctuation that allows the letting-go to happen.

The Ending work is not emotional indulgence. It is a structural requirement of the transition. Skipping it produces a workforce that carries the old role into the new assignment, producing hybrid behaviour that is neither fully the old nor fully the new.

Designing for the Neutral Zone

The Neutral Zone is the longest phase and the one most likely to be experienced as failure by the sponsor. The population is uncomfortable, productivity is down, complaints are up, confidence is shaky. The sponsor’s instinct is to push through — to accelerate the timeline, to add more training, to intensify the communications. The instinct is wrong.

Neutral Zone interventions protect the ambiguity rather than fight it:

• Accept lowered productivity temporarily. A 15–25% productivity dip for 60–120 days is normal during a substantive role change. A sponsor or manager who interprets the dip as a programme failure and escalates pressure will produce either regression to the old behaviour or performance theatre.
• Enable experimentation. The Neutral Zone is when new behaviours are tried. A culture that punishes early failures in the new way of working will suppress the experimentation the transition needs. Managers are trained (Article 28) to coach through early failures rather than grade them.
• Provide psychological structure without closure. Regular check-ins with managers and peers, support groups (formal or informal), access to coaching. The check-ins do not resolve the ambiguity — that would be premature — they accompany the learner through it.
• Communicate that the discomfort is the process, not a malfunction. A workforce that is told “this discomfort means the programme is working” will carry the Neutral Zone differently from one that is told “everything is going smoothly, please continue.”

The expert’s discipline at the Neutral Zone is patience. Programmes that drive through the Neutral Zone with intensified pressure produce worse outcomes than programmes that create space for it.

Designing for the New Beginning

The New Beginning is often assumed rather than resourced. The organisation’s implicit belief is that once the Neutral Zone passes, employees will commit naturally. Bridges showed that this is not reliable. Commitment requires energy, and the energy must come from somewhere.

Three design elements support the New Beginning:

• A clear picture of the new. Employees commit to a specific picture, not to a general vision. The specific picture — “this is what your work week looks like in the new role, with real examples” — is the object of commitment.
• Early small wins for each individual. A learner in the new role needs early, small, personal wins that prove to them (not to the sponsor) that the new role works for them. Wins that are claimed on their behalf by leadership but that they did not experience personally do not build commitment.
• Symbolic markers of arrival. Where the Ending was marked ritually, the New Beginning is marked similarly. The new role gets a new title, a new team membership, a new visible anchor. These are small but material; they signal that the transition has completed.

The expert’s design test: six months into the new role, do individual contributors describe themselves using the new role’s identity, or do they still describe themselves using the old? If the former, the New Beginning has landed. If the latter, the commitment work was under-resourced.

Common failure: announcing as if the transition were automatic

The recurring failure in AI role changes is the announcement treated as the whole change. The email goes out, the all-hands is held, the new role descriptions are published, and the sponsor assumes the transition is now under way. Three months later, adoption metrics are flat, resistance is rising, and the sponsor is confused about what happened.

What happened is that the organisation announced a change and did not design the transition. The population is somewhere — mostly in the Ending phase, still letting go — and there is no structural support for their work. The manager cadence is generic rather than transition-aware; the communications are about the new without acknowledging the loss; the performance system is being applied as if the new role were fully embedded when in fact the population is mid-transition.

The remedy is a transition plan as a first-class artefact, distinct from the change plan. The change plan covers the external event. The transition plan covers the psychological process: which populations are at which phase, what their structural needs are, how managers are supporting them, when the phases are expected to shift, what signals trigger plan adjustment. The transition plan is typically a 6–18 month artefact; the change plan is often shorter.

Two real-world anchors

Bridges’s case material — work-redesign transitions outside AI

Bridges’s 4th edition (2017) documents a range of work-redesign transitions — from manufacturing automation of the 1990s through services restructuring in the 2000s and early 2010s — and the patterns that predict successful versus unsuccessful transitions. The patterns Bridges identifies are technology-agnostic: the Ending needs to be named and honoured, the Neutral Zone needs to be protected rather than pushed through, the New Beginning needs to be energised. These patterns map directly onto AI role changes with no meaningful adjustment. Source: https://wmbridges.com/resources/ and William Bridges Associates publications.

The lesson: the Bridges framework is not specific to AI, and its long track record across other transitions gives the expert confidence in applying it to AI transitions. The patterns predict; the underlying psychology is stable.

The Dutch public-sector transitions following the Toeslagenaffaire

The Dutch government’s reckoning with the Toeslagenaffaire (child-benefits scandal) between 2020 and 2025 has included role-redesign at scale in the tax administration (Belastingdienst). Public parliamentary inquiry reports and Autoriteit Persoonsgegevens decisions document the institutional failure and the subsequent restructuring of roles, responsibilities, and oversight. Source: https://www.tweedekamer.nl/kamerstukken/detail?id=2020D53175 and Autoriteit Persoonsgegevens publications.

The transition work the Dutch civil service is conducting — covered more fully in the Case Study for this credential — is a long-running example of what happens when the Ending (of the old algorithmic risk-scoring practice) is not adequately honoured, the Neutral Zone (of rebuilding professional legitimacy) stretches into years, and the New Beginning (of human-centred decision-making) remains uncertain. The case demonstrates the costs of treating a transition as an announcement rather than as a structured psychological process. The lesson for the enterprise expert: the pattern is the same at organisational scale as at national scale.

Learning outcomes — confirm

A learner completing this article should be able to:

• Distinguish change (the external event) from transition (the psychological process) and apply the distinction to a described AI role redesign.
• Name the three Bridges phases and design structural support for each.
• Argue why the Ending phase is the most under-designed and what concrete design elements support it.
• Defend the Neutral Zone from the sponsor’s instinct to accelerate through it.
• Resource the New Beginning through specific pictures, personal wins, and symbolic markers.
• Produce a transition plan as a first-class artefact alongside the change plan.

Cross-references

• EATF-Level-1/M1.6-Art08-Workforce-Redesign-and-Human-AI-Collaboration.md — Core Stream workforce-redesign anchor.
• Article 18 of this credential — Bridges’s place in the choice framework.
• Article 24 of this credential — task-level decomposition (input to role redesign).
• Article 25 of this credential — redesigned role specification.
• Article 26 of this credential — redundancy planning (loss-laden transitions).
• Article 28 of this credential — manager enablement (structural support during transition).

Diagrams

• Timeline — Bridges three phases across an 18-month AI role-change transition, with population distribution per phase at each time slice.
• Matrix — phase × leader action × manager action × individual contributor signal, with intervention class per cell.

Quality rubric — self-assessment