Coaching for AI Adoption: Building Leadership Capability for Organisational Transformation

AI adoption is advancing faster than many organisations' ability to redesign work around it. The 2026 Stanford AI Index reports that 88% of surveyed organisations were using AI in at least one business function in 2025, while generative AI was in use in at least one function at 70% of organisations. Yet adoption alone tells senior leaders relatively little about whether AI is improving decisions, changing workflows or creating sustainable organisational capability.

That distinction matters because AI implementation ultimately changes human work. Leaders must determine which decisions can be augmented, which processes should be redesigned, where human judgement must remain decisive and how employees should work differently as AI becomes embedded in daily activity. OECD analysis of AI adoption in firms similarly identifies skills, organisational capabilities, data, infrastructure and management practices among the conditions influencing successful adoption.

For Chief People Officers, Talent leaders, Learning and Development functions, Organisational Development specialists and Transformation Directors, the core challenge is therefore behavioural as well as technological. Organisations need people who can use AI competently, question it appropriately, recognise its limitations, apply judgement and continually adapt as roles and workflows evolve.

Coaching for AI adoption can support that transition. Within a wider programme of strategy, technology, governance, capability building, organisational design and change management, Coaching for Organisational Transformation can help leaders and employees translate AI ambition into new ways of thinking, deciding and working.

AI adoption becomes valuable when work and behaviour change

Providing employees with an AI tool does not determine how effectively they will use it.

The same technology can produce markedly different outcomes depending on the task, the employee's existing expertise and the way AI is incorporated into work. In one influential field study involving 5,179 customer-support employees, access to a generative AI assistant increased productivity by approximately 14% on average, with substantially larger gains among less experienced and lower-performing workers. More experienced employees experienced much smaller effects.

The implication is important for organisational buyers. AI adoption cannot be managed solely through licences, technical training and adoption metrics because value emerges from interactions between technology, task design and human capability.

Current McKinsey research makes a similar distinction. Its 2026 analysis of AI transformation argues that organisations moving furthest towards business impact are redesigning workflows and investing in behaviours, skills, leadership practices and organisational change alongside technology deployment.

This moves the conversation from "How many people are using AI?" towards more demanding questions.

Are people using AI for the right tasks? Do they know when to challenge its output? Are managers redesigning work intelligently? Are teams sharing effective practices? Can employees distinguish productive augmentation from unnecessary automation? Are leaders modelling responsible experimentation while maintaining appropriate accountability?

These are leadership and behavioural questions.

Coaching can provide a structured mechanism for working through them.

AI adoption creates a new leadership capability requirement

Leaders do not need to become machine-learning engineers to lead AI-enabled organisations. They do need sufficient understanding to make sound decisions about work, capability and risk.

This requirement is already becoming visible in labour-market evidence. The OECD's 2026 review of AI and skills identifies skills shortages as a significant constraint on AI adoption and notes that organisations increasingly require employees who can use, analyse and interpret information produced through AI-enabled systems. The report also finds that workers receiving AI-related training are more likely to report positive effects on performance and working conditions.

Technical literacy is only part of the equation. The World Economic Forum's Future of Jobs Report 2025, based on responses from more than 1,000 employers, identifies AI and big data among the fastest-growing skill requirements while also highlighting analytical thinking, resilience, leadership and collaboration as continuing core capabilities. Seventy-seven per cent of surveyed employers expected to prioritise reskilling and upskilling their workforce in response to AI.

Senior leaders therefore need to develop several capabilities simultaneously: sufficient AI literacy to understand opportunities and constraints; systems thinking to anticipate consequences across workflows and roles; decision quality under uncertainty; and the behavioural adaptability required to change established ways of working.

Leadership coaching can help make these requirements concrete.

A coach might work with a senior executive who recognises AI's strategic importance but continues to approve processes based on assumptions developed before AI became available. Another leader may be adopting AI rapidly without adequately interrogating data provenance, decision accountability or employee consequences. A functional leader may understand AI intellectually but struggle to translate that understanding into new expectations for managers.

Evidence-based coaching creates space to examine these patterns and convert insight into deliberate leadership behaviour.

A 2023 meta-analysis of randomised controlled trials found a moderate positive overall effect of executive coaching, with stronger effects on behavioural outcomes than on attitudes or personal characteristics. That evidence does not establish that coaching causes successful AI adoption, but it supports coaching's relevance where the organisational requirement is sustained behavioural adaptation.

AI adoption requires experimentation without abandoning judgement

Many organisations want employees to experiment with AI while simultaneously expecting responsible use. Those objectives can create tension.

If employees fear making mistakes, experimentation becomes superficial. If experimentation occurs without clear boundaries, the organisation may create privacy, quality, intellectual-property, ethical or decision-accountability risks.

Senior leaders therefore need to establish an environment in which employees can learn while understanding where caution is required.

The growing use of algorithmic systems in managerial decisions illustrates the complexity. OECD research across six countries found that managers often perceive algorithmic-management technologies as improving decision quality, yet concerns persist around accountability, interpretability and worker protection. The same research highlights worker consultation as one mechanism for improving acceptance and reducing implementation risk.

Coaching can support leaders in navigating these judgement calls because it encourages reflection without prescribing universal answers.

A manager might use coaching to examine:

• which parts of a decision should legitimately be delegated to AI;
• what evidence would justify challenging an AI-generated recommendation;
• how an employee should be involved when AI materially affects their work;
• where efficiency goals could unintentionally reduce learning or capability;
• how to communicate uncertainty without undermining confidence.

The value lies in improving the quality of the leader's reasoning and subsequent behaviour.

This becomes particularly important as organisations move beyond individual productivity tools towards AI-enabled workflows, automation and agentic systems, where choices made by leaders can affect roles, decision rights and organisational structure.

Training builds knowledge; coaching helps translate knowledge into practice

Formal learning has an essential role in AI adoption.

Employees need instruction on approved tools, information security, prompt design, appropriate use cases, data handling and organisational policy. Managers need additional understanding of governance, workflow design and the implications of AI for roles and performance.

Coaching complements this activity by focusing on application.

An employee may complete AI training successfully but continue using established processes because those processes feel safer. A manager may understand an AI tool but struggle to identify where a workflow should change. An executive may know the organisation's responsible-AI policy but still require support when balancing competing commercial, ethical and workforce considerations.

These situations illustrate the gap between knowing and doing.

Workplace coaching evidence suggests that structured coaching interventions can positively influence organisationally relevant outcomes, while also showing considerable variation across contexts and programme designs. A 2023 meta-analysis therefore argues for stronger measurement and greater attention to the conditions under which coaching is deployed.

For AI transformation, this means coaching should be aligned with defined capability and behavioural outcomes.

It should reinforce learning programmes, workflow redesign and manager expectations, not operate as an isolated development activity.

AI adoption should be designed around different employee populations

An organisation-wide AI strategy does not imply identical development needs across the workforce.

Executives may need coaching around strategy, governance, organisational design and investment decisions. Functional leaders may need support identifying opportunities to redesign workflows. People managers may require help leading teams whose roles are changing. Employees may need practical support developing confidence and judgement in everyday AI use.

Some populations will also experience greater disruption than others. The International Labour Organization's updated analysis of generative AI exposure emphasises that AI's impact occurs at task level and varies significantly across occupations, reinforcing the need to consider job transformation more carefully than broad predictions about whole roles disappearing.

This has practical implications for Talent and L&D leaders.

A useful coaching architecture might combine executive coaching for strategic decision-makers, leadership coaching for managers implementing new operating practices, team coaching where entire workflows are being redesigned and group coaching where cohorts face similar adoption challenges.

The developmental question should determine the modality.

Human coaching and AI coaching can reinforce AI adoption in different ways

AI adoption also creates an unusual opportunity: AI itself can become part of the developmental infrastructure supporting the transformation.

AI coaching can provide employees with accessible opportunities to reflect, prepare and practise during the working day. A manager could use an AI coach before a conversation about changing responsibilities, explore different approaches to implementing an AI-enabled workflow or reflect on whether they are relying too heavily on automated recommendations.

This frequency matters because behavioural change develops through repeated application, feedback and reflection.

Human coaches provide complementary capabilities. They can work with ambiguity over time, identify patterns across different situations, bring relational depth and challenge leaders where assumptions, emotions or organisational politics are shaping decisions.

The appropriate model therefore combines different forms of developmental support according to context.

At BOLDLY, AI coaching through momentLeader extends opportunities for reflection and development into the flow of work, while evidence-based human coaches support situations requiring greater relational depth, contextual judgement and sustained developmental challenge.

AI coaching should operate within the same disciplined governance philosophy expected of enterprise human coaching. Access, privacy, appropriate use, escalation pathways and quality assurance all matter.

Coaching culture can accelerate organisational learning around AI

AI adoption is unusually dynamic because the technology and its applications continue to change after implementation.

The 2026 Stanford AI Index documents both rapid organisational adoption and continuing advances in AI capability. Organisations therefore cannot assume that a single training cycle will establish the capabilities they need indefinitely.

A coaching culture can strengthen the organisation's capacity for continuous adaptation.

Managers who routinely ask employees what they are learning, what assumptions they are testing and where AI is helping or hindering performance create conditions for knowledge to circulate. Teams can compare approaches, surface unexpected risks and identify emerging practices instead of waiting for central functions to codify every development.

This does not reduce the need for formal governance. It strengthens the learning system around it.

For L&D and Organisational Development leaders, that creates an important opportunity. AI capability can become part of everyday leadership and management practice instead of remaining a specialist digital-skills initiative.

Coaching operations become critical when AI development moves to enterprise scale

When AI adoption affects thousands of employees across markets, functions and leadership levels, coaching itself requires operating discipline.

Enterprise coaching programmes need quality assurance, appropriate coach selection, secure workflows, clear confidentiality arrangements, scalable administration and measurement capable of generating useful organisational insight without exposing private coaching conversations.

Technology-enabled coaching operations enable an organisation to coordinate these components while retaining appropriate governance.

Measurement should also move beyond utilisation.

Relevant questions include whether leaders are changing targeted behaviours, whether teams are adopting redesigned workflows, whether employees are developing confidence and appropriate judgement and whether coaching insights reveal systemic obstacles that the transformation programme needs to address.

Attribution should remain measured. Productivity, adoption, retention and organisational performance depend on multiple interacting factors, including technology quality, process design, incentives, leadership, skills, data and operating context.

Coaching contributes to this system by strengthening the human capability required to operate it.

Frequently asked questions

What is coaching for AI adoption?

Coaching for AI adoption is developmental support designed to help leaders and employees adapt their decisions, behaviours and ways of working as AI becomes embedded in the organisation. It can include executive coaching, leadership coaching, team or group coaching and appropriately governed AI coaching.

How is coaching different from AI training?

Training primarily builds knowledge and technical capability. Coaching focuses on how an individual or team applies that knowledge in real situations, including judgement, behaviour, decision-making and adaptation. Organisations typically need both.

Can coaching increase AI adoption?

Coaching may contribute to adoption by helping employees address behavioural barriers, develop confidence and integrate new practices into their work. Adoption also depends on technology quality, leadership, incentives, skills, workflow design, governance and organisational readiness, so coaching should form part of a broader transformation approach.

Should every employee receive a human AI coach?

Not necessarily. Human coaching is particularly valuable where leadership complexity, judgement, role transition or organisational consequences justify deeper developmental support. AI coaching, group coaching and manager-led coaching conversations can extend development to broader populations where appropriate.

How should organisations measure coaching for AI adoption?

Measurement should connect coaching objectives with relevant behavioural and organisational indicators. These may include changes in leadership behaviour, confidence and judgement, adoption of redesigned workflows, manager effectiveness or capability development. Usage and satisfaction are useful operational indicators but provide an incomplete picture of organisational impact.

BOLDLY: Coaching for Organisational Transformation in the age of AI

AI adoption is ultimately an organisational transformation challenge. Technology creates possibilities, but leaders, teams and organisational systems determine how those possibilities translate into performance, capability and sustainable ways of working.

BOLDLY helps organisations build the human capability required for that transition through an integrated Coaching for Organisational Transformation ecosystem.

This brings together evidence-based human coaching, behavioural science and adult development, AI coaching through momentLeader, technology-enabled coaching operations, coaching governance and quality assurance, screened and vetted coaches and global delivery capability.

Combined with Bendelta organisational transformation expertise, this approach enables organisations to connect individual development with the wider leadership, behavioural and organisational requirements of AI transformation.

The starting point is a practical organisational question: where does AI require people to think, decide, lead or work differently?

That is where coaching can make its most useful contribution.

About the Author: Maggie Wong

Maggie Wong is a Coach Business Partner at BOLDLY, where she helps bring world-class coaching to organizations around the globe. She supports the onboarding and development of new coaches while connecting clients with exceptional coaching talent tailored to their needs. Maggie works closely with leaders and organizations on initiatives related to top team effectiveness, leadership development, and large-scale coaching programs. Passionate about enabling meaningful growth, she focuses on building strong partnerships that help individuals, teams, and organizations thrive.