AI Coaching: What P&C Leaders Need to Know About the Human Element
Two recent papers by Professor Tatiana Bachkirova at Oxford Brookes University have tackled a crucial question in our field: can AI replicate human coaching? Her research, along with discussions I’ve had with colleagues in the Coaching Psychology Special Interest Group of the New Zealand Psychological Society, reveals some fascinating insights about the future of coaching.
Why AI Can’t Replace Human Coaches – And That’s Not Going to Change
As a leadership coach, facilitator and programme designer, I’ve noticed an interesting pattern. Every time I speak with People and culture leaders about coaching, the conversation inevitably turns to AI. Will AI replace human coaches? Should we invest in AI coaching platforms? Will we fall behind if we don’t?
These are thoughtful questions, driven by very real pressures to scale development opportunities and manage costs. But they reflect a fundamental misunderstanding about what coaching actually is.
Here’s the key insight: The limitations of AI in coaching aren’t about current technology. They’re about the inherently human nature of coaching itself. And this won’t change, no matter how sophisticated AI becomes.
AI cannot, in principle, provide a stand-alone coaching service that is equivalent to human coaching.
Why This Matters Now
You might be feeling pressure to adopt AI coaching solutions. Vendors are making compelling pitches about democratising coaching through technology. The promise is appealing: wider access, lower costs, and consistent quality.
But before making significant investments, it’s crucial to understand what makes coaching work in the first place. Tatiana Bachkirova and Rob Kemp published a paper earlier this year proposing six essential characteristics of effective coaching – and they all require real human intelligence.
The Six Elements That Make Coaching Work
1. Joint Inquiry
Coaching isn’t about following a script or applying pre-set solutions. It’s about two humans exploring challenges together, drawing on shared experiences and perspectives. When a leader is grappling with a complex challenge, the coach’s lived experience and ability to grasp nuanced meanings make all the difference.
2. Making Sense of Experience with Action Focus
Human coaches bring their embodied understanding of organisational life. They can help leaders interpret situations holistically, considering emotions, politics, and cultural context. AI can process information, but it can’t truly understand what it feels like to navigate workplace dynamics.
Our cognition is embodied: it happens not just in the brain. Emotions, for example, as an aspect of the whole body, are central to our thinking and action
It is our body that grasps the gestalt and then acts. Humans respond to the world immediately rather than through the analysis of details. In grasping the complexity of clients’ situations, this is how coaches identify patterns to reflect back to clients and to facilitate further exploration. AI cannot do this.
3. Value-Based Approach
Effective coaching requires ethical judgment and genuine care for the client’s development. While AI can be programmed with ethical guidelines, it can’t actually hold values or make nuanced moral judgments in complex situations.
4. Contextual Understanding
Every organisation has its unique culture, politics, and unwritten rules. Human coaches draw on their experience to help leaders navigate these waters. AI, lacking real-world experience, can’t grasp these subtle but crucial dynamics.
5. Trust-Based Relationships
The research is clear: the coaching relationship is the biggest predictor of coaching success. This requires genuine empathy, authentic connection, and the ability to build real trust – all uniquely human capabilities.
6. Contracting-Based Engagement
Effective coaching requires ongoing negotiation between coach, client, and organisation. This includes handling sensitive issues, managing confidentiality, and adapting the approach as needs change – complex human interactions that AI cannot authentically replicate.
Where AI Can Actually Help
While AI can’t replace human coaching, it can play a valuable supporting role. Here’s a balanced view of where technology can enhance coaching programmes:
Pre and Post-Session Support
Digital reflection tools for clients to prepare for sessions: AI can provide structured prompts and questions to help clients reflect deeply before coaching conversations. For example, a tool might guide a leader through examining recent challenges, identifying patterns in their responses, and clarifying what they want to explore with their coach.
AI-driven prompts for action plan follow-up: Intelligent systems can send contextually relevant reminders and check-ins about committed actions. Rather than generic reminders, these can reference specific goals and adapt timing based on the client’s engagement patterns.
Progress tracking and milestone celebrations: AI tools can help clients monitor their development journey, visualising progress over time and automatically highlighting achievements. This might include tracking behaviour changes, logging reflections, or measuring progress against specific leadership competencies.
Resource recommendations based on discussion themes: Using natural language processing, AI can analyse coaching session notes (with appropriate permissions) to suggest relevant articles, videos, or exercises that align with the client’s development focus areas.
Scale and Accessibility
Basic skill practice through simulations: AI can create safe spaces for practising specific skills like difficult conversations or presentation techniques. These tools can provide immediate feedback on aspects like language choice, pace, or structure.
On-demand learning resources: AI can curate and deliver personalised learning content based on the client’s development goals, learning style, and previous engagement patterns. This ensures relevant support is available between coaching sessions.
Guided reflection exercises: Interactive tools can lead clients through structured reflection processes, helping them unpack experiences and extract learning in between human coaching sessions. These might use branching scenarios based on responses.
Quick reference tools for frameworks and models: AI can provide contextual reminders of coaching frameworks and models at relevant moments, helping clients apply their learning in real-world situations.
Coach Support Tools
Session summary generation: AI can help coaches create structured session notes, identifying key themes and patterns while leaving coaches free to focus fully on the client during sessions. This could include highlighting potential areas for follow-up.
Pattern recognition across multiple clients: With appropriate privacy safeguards, AI can help coaches identify common themes or challenges across their client base, informing their professional development and resource preparation.
Administrative task automation: AI can handle scheduling, reminder management, and basic documentation, freeing coaches to focus on the human elements of their work. This might include automated pre-session questionnaires or feedback collection.
Evidence-based intervention suggestions: AI can support coaches with real-time suggestions of relevant coaching interventions based on client responses and current research, while leaving the crucial decisions about if and how to use these insights to the coach’s professional judgment.
The key is using AI to augment human coaching, not replace it.
Making Smart Decisions About Coaching Technology
For P&C leaders, this suggests a clear framework for decision-making:
1. Be clear about what coaching is trying to achieve in your organisation
Before evaluating any technology, establish your coaching strategy’s core purpose. Are you aiming to develop senior leaders, support succession planning, enable transformation, or build specific capabilities? Different goals require different approaches. For example, supporting leaders through organisational change typically requires deep human engagement, while skill development might benefit from a blended approach. Map out your objectives, target audience, and desired outcomes before considering technology solutions.
2. Understand which elements require human intelligence
Review Bachkirova and Kemp’s six essential elements of coaching discussed earlier:
- Joint inquiry requires human presence for genuine exploration
- Making sense of experience needs embodied understanding
- Value-based approaches depend on human ethical judgment
- Contextual understanding draws on lived organisational experience
- Trust-based relationships need authentic human connection
- Contracting-based engagement requires real-time negotiation
Evaluate your coaching needs against these elements. Where are these human capabilities most crucial for your desired outcomes? Which aspects might be augmented (but not replaced) by technology?
3. Look for AI tools that support rather than replace human coaches
Seek solutions that enhance rather than eliminate human coaching relationships. Good AI tools should:
- Facilitate preparation and reflection between sessions
- Provide structured frameworks for practice
- Help track progress and commitments
- Support coaches with administrative tasks
- Offer relevant resources and exercises
- Enable scalable reinforcement of coaching conversations
4. Be wary of vendors promising AI can replicate human coaching
Apply critical thinking to vendor claims. Ask specific questions about:
- How the AI was trained and validated
- What research supports their effectiveness claims
- How they handle complex emotional situations
- Their approach to ethics and privacy
- Their understanding of coaching psychology
- Their protocols for detecting and responding to mental health concerns,
- Their definition of “coaching” vs information delivery
Remember: if it sounds too good to be true, it probably is.
5. Consider how to use technology to extend the impact of human coaching
Think creatively about leveraging technology to:
- Prepare clients for coaching through guided reflection
- Reinforce learning between sessions
- Scale insights from coaching across the organisation
- Create communities of practice
- Measure and demonstrate impact
- Support ongoing development after formal coaching ends
Look for ways to create sustainable development ecosystems where technology and human coaching complement each other rather than compete.
Ethical Implementation Guidelines
When implementing AI coaching tools, consider:
Data Privacy and Security
- Where is data stored?
- Who has access?
- How is confidentiality maintained?
Cultural Safety
- Does it acknowledge different cultural perspectives?
- Can it adapt to diverse communication styles?
Transparency
- Are users clear about when they’re interacting with AI?
- Is the scope of AI support clearly defined?
- How are limitations communicated?
We arrive at moral evaluations not through solving logical puzzles but through consideration of what is irreducible in us: subjectivity, dignity, desire – all the things that AI doesn’t have. This subjectivity is an important player when meaning is co-constructed through relationship with others, and this is why in coaching we deal with meaning tentatively and together as it emerges in relationship with clients.
A Te Ao Māori Perspective
As a Pākehā coach, I offer my understanding of how Te Ao Māori principles illuminate the inherently human nature of coaching, while acknowledging that my perspective comes from my ongoing journey of learning and understanding. From my perspective, the concept of whanaungatanga reminds us that genuine relationships and connections are at the heart of development. Similarly, tuakana-teina relationships – where an experienced person guides a less experienced one – are built on mana-enhancing interactions that go far beyond simple knowledge transfer.
The principle of ako – the interchangeable nature of teaching and learning – reflects how effective coaching involves mutual growth and understanding. When we view coaching through this lens, it becomes clear that AI, lacking mauri (life force) and unable to engage in genuine whanaungatanga, cannot replicate these essential human elements.
Moving Forward
The future of coaching isn’t about replacing humans with AI. It’s about thoughtfully combining human intelligence with technological tools to create better development outcomes.
The next time you’re evaluating coaching investments, start with this question: What aspects of this solution require real human intelligence, and what can be effectively augmented by technology? How will this honour our cultural values and obligations? What’s the right balance for our organisation?
This clarity will help you make better decisions about where to invest – and help you explain those decisions to stakeholders who might be dazzled by AI’s promises.
What’s your experience with AI coaching tools? I’d be interested to hear how you’re thinking about these issues in your organisation.
The coaching papers that inspired this article:
Tatiana Bachkirova & Rob Kemp (27 Jun 2024): AI coaching: democratising coaching service or offering an ersatz?, Coaching: An International Journal of Theory, Research and Practice, DOI: 10.1080/17521882.2024.2368598
https://doi.org/10.1080/17521882.2024.2368598
Tatiana Bachkirova (November 2024): Why Coaching Needs Real Intelligence, Not Artificial Intelligence, Philosophy of Coaching: An International Journal Vol. 9, Issue 2, November 2024, 6-15 http://dx.doi.org/10.22316/poc/09.2.02
Coaching Psychology Special Interest Group – LinkedIn page
About the author: I am an executive coach with a Master of Science in cognitive behavioural coaching. My approach integrates adult development theories, embodied leadership principles, mindfulness-based leadership, complex adaptive leadership, and motivational approaches.
