Can I use Ai in coaching?

Listen to a summary of this article here:

A complete guide including a look into the future of AI in coaching.

Let’s be candid. For many coaches, the rise of Artificial Intelligence feels like a threat hiding in plain sight. News cycles showcase chatbots writing articles, tools doing legal review, and algorithms nudging behaviour in ways that once needed human hands. If you’ve invested years in training and your reputation rests on trust and presence, the fear is understandable: will AI in coaching undercut fees, divert clients to cheaper tools, or even make coaches redundant? That anxiety is not trivial; it reflects real shifts in discovery and delivery that are reshaping expectations across the industry. Yet panic helps no one. The better question is how to adapt, because AI is already here and increasingly part of how clients find, evaluate, and experience coaching.

The truth is that coaching is different from transactional services. At its best it is deeply relational, built on empathy, contextual judgement and ethical boundaries. These qualities don’t vanish because a model can draft a tidy paragraph or produce a neat reflection prompt. What changes is the environment around the craft. AI in coaching alters the routes by which clients discover you, the scaffolding that supports practice between sessions, and the expectations organisations bring to scale and measurement. Coaches who embrace this with discernment will keep the human work at the centre while letting technology handle the repetitive, the structured and the scalable. Those who opt out entirely may find visibility and consistency harder to sustain as peers adopt new capabilities.

Discovery has shifted online. Where referrals and conferences once dominated, executives and professionals now meet coaches via search, short-form social content, podcasts and webinars. That change matters because AI in coaching doesn’t just influence delivery; it shapes discovery too. Platforms and media now use recommendation systems to surface voices, while employers increasingly pilot digital coaching at scale before selecting individual coaches. The practical implication is simple: visibility is a business asset. If your ideas, case studies and perspective don’t show up where buyers are looking, they often won’t be seen at all — even when your coaching is outstanding. Studies and reporting over the last two years detail this move towards digital-first encounters.

This “discovery shift” sits alongside a structural shift in the market. A layered landscape is forming: low-cost, scalable services; hybrid offerings that blend human support with automation; and premium, transformational human coaching at the top. The trend is visible in enterprise platforms: BetterUp launched AI in coaching features to broaden access; CoachHub introduced AIMY/Companion for 24/7 support; and providers like Ezra explain where AI augments matching and readiness rather than replacing coaches. For independents, that layered market is not doom; it’s an invitation to clarify positioning. Let AI scale routine support while your human work focuses on judgement, trust and impact.

It’s already in production. In January 2025, BetterUp announced AI coaching capabilities aimed at giving more employees structured, contextual support “in the flow of work”, accompanied by adoption figures from early customers. Business Insider reported rapid uptake, with a majority of users preferring hybrid models that combine human and AI support — a signal that most clients don’t want bots instead of people; they want better access and continuity around coaching. Meanwhile, CoachHub has publicised conversational features and an AI Companion integrated into Microsoft Teams for 24/7 nudges and reflection. This is not a prototype moment; it is live software across global employers.

ICF has moved in step with this reality. In 2024, the International Coaching Federation published its AI in coaching Framework and Standards to help the field adopt technology responsibly. Crucially, the standards don’t treat AI as monolithic. They classify applications by how they’re used and the risks they pose, recognising that scheduling reminders are not the same as a conversational bot simulating coaching. The standards aim to protect client welfare, clarify boundaries between coaching and therapy, and guide providers and purchasers toward systems that are ethical, accessible and evidence-informed. For working coaches, that guidance is a guardrail: embrace tools, but keep practice human-centred and transparent.

The ICF’s documents outline a pragmatic taxonomy (see table). Coach-assisting tools sit in two groups: scheduling (reminders, planning, journals) and data processing (summaries, sentiment analysis, readiness indicators). These typically support human coaches and carry lower risk. Coaching-service tools sit in two higher-risk groups: interactive (nudges, quizzes, commitments) and conversational (chatbots, voicebots, avatars) that engage directly with clients. As systems move from assistive to service-delivery, the standards call for stronger safeguards, because AI in coaching at that level processes more sensitive topics and can shape behaviour in real time. The rationale is straightforward: more impact means more responsibility.

The ICF emphasises several themes: transparency about where and how AI is used; informed consent for data processing; conscientious handling of coach and client data; minimising bias; and clear boundaries separating coaching from therapy. The standards also encourage providers to self-certify or use third-party evaluation against competency and ethics criteria, rather than taking an “all or nothing” view of compliance. For coaches, this is a template for governance. For buyers, it’s a way to ask better questions. And for developers, it’s the field saying, “build responsibly so AI in coaching extends access without eroding trust.”

It’s the most common fear — and it’s not what current evidence suggests. Reporting in the Financial Times shows employers piloting AI “coachbots” for rehearsal, reflection and guidance on everyday management problems, precisely because these tools are available on demand and reduce the barrier to getting help. But those programmemes don’t eliminate the need for humans. They triage routine needs and widen the funnel, while complex or sensitive issues continue to flow to qualified coaches. As one FT series noted, AI in coaching is expanding access for underserved groups, not ending premium human work. The early signal: automation at the base, premium at the top.

Adoption data outside coaching points in the same direction. Business Insider’s reporting on BetterUp’s AI launch cited high satisfaction among early users and a preference for hybrid support — humans plus AI — rather than AI alone. Multiple enterprise platforms now position their offerings explicitly as blended ecosystems. That’s consistent with ICF’s stance that conversational systems carry higher risk and must be deployed with safeguards and clear boundaries. In other words, what’s scaling fastest is not “bot instead of coach,” but “coach plus infrastructure,” with AI in coaching making continuity, nudges and measurement easier while the human relationship remains the centre of gravity.

If you’re new to this, begin with assistive tasks that save time without touching clinical boundaries. Use a writing model to generate first-draft newsletters, blog outlines and show notes in your own voice; then edit for nuance. Summarise call notes to spot themes your client cares about and to create hand-crafted reflection prompts between sessions. Translate accessibility into action by offering transcripts and plain-English session summaries. These small shifts compound. Over a quarter, AI in coaching can recover hours each week and lift perceived service quality because your post-session follow-through is consistent, readable and timely — without you spending your evenings on admin.

Next, use models to test messaging. Feed in anonymised audience language from testimonials and discovery calls to surface phrases that resonate; then A/B test headlines and CTAs across emails and LinkedIn posts. This is where AI in coaching meets marketing discipline. You’re not outsourcing your voice; you’re using data to reduce guesswork. As you grow comfortable, add light client-facing elements with consent: goal-tracking nudges, pre-session check-ins, or structured journalling that scaffolds behaviour change. Keep the human relationship central by adjusting those prompts in session, reinforcing accountability in a way no generic prompt can. That pairing — human depth plus digital cadence — is powerful.

One of the hardest tasks as a coach is articulating a niche that is both authentic and commercially viable. Start by mining your own data. Ask a model to review anonymised testimonials, intake forms and session notes to identify recurring goals, obstacles and outcomes. You’ll see patterns you “know” intuitively, but with sharper language. From there, AI in coaching can draft positioning statements, service pages and FAQs expressed in words clients actually use. This isn’t cosmetic. When your language matches their search intent, discovery improves and consultations feel warmer because clients feel understood before you’ve met.

Then pressure-test demand. Use AI to map your niche to adjacent pain points across roles and industries. For instance, “mid-career leadership reset” might connect to performance conversations, influence without authority and values-based decision-making. Build a content sequence that addresses those edges with short, useful pieces. Your aim is to become findable where your clients already spend attention. With AI in coaching handling first drafts and repurposing long-form pieces into clips, posts and emails, you stay consistent without burning out. Consistency, not viral spikes, is what compounds reach — and it’s where AI quietly levels the playing field for independents.

Expect downward pressure at the entry level and rising willingness to pay at the top. As automated tools handle routine reflection and micro-nudges at low cost, buyers will reserve premium spend for work that clearly requires human judgement, confidentiality and presence — the terrain of culture change, leadership transitions and identity shifts. The market data supports this layered view. The ICF values the global industry at $4.56bn in 2023 with strong growth projections (see chart), while mainstream outlets track rapid enterprise uptake of digital platforms that complement, not replace, human coaching. In short, AI in coaching expands access; human coaches anchor transformation.

For independents, ROI improves when three elements converge: credibility, positioning and a consistent business engine. ICF-aligned training signals standards and ethics to buyers; a clear niche sharpens relevance; and repeatable growth rhythms — content, outreach, client stewardship — keep the pipeline moving. Here AI in coaching is a force multiplier. It doesn’t decide fees; it protects margins by reclaiming hours and smoothing delivery, letting you spend more time on premium work while still maintaining visibility and client experience. The coaches who combine rigorous practice with simple, tech-supported operations will see the clearest ROI uplift as the market stratifies.

Bias, privacy and boundary-blur are the big three. Models trained on partial or skewed datasets can reinforce stereotypes unless monitored and corrected. That’s not an abstract problem; it shows up in the questions prompts ask, the examples they surface, and the assumptions they make about “good” behaviour. The ICF standards highlight the need to minimise bias through testing and oversight. Privacy is equally non-negotiable. AI in coaching often processes sensitive personal data. Coaches should know where data is stored, how it’s used, and how clients can opt out — and should explain this simply. If you wouldn’t be comfortable reading the policy to a client, it’s not ready.

The third risk is boundary-blur. Conversational systems can feel empathic while lacking true attunement. It’s easy for a client to over-disclose or for a tool to drift into territory more appropriate for therapy. The ICF explicitly warns against this, urging clear distinctions in marketing and practice. In short, treat AI in coaching as an assistant that scaffolds behaviour change and reflection, not a replacement for human presence. When in doubt, dial back client-facing use and strengthen your assistive automations instead. The goal is to enhance safety and effectiveness, not to test the edge of what a chatbot can do.

You don’t need an in-house legal team to act like a pro. Start with a plain-English data statement covering tools used, data retained, how clients can opt out, and how you handle transcripts and summaries. Then create a short “AI use” explainer that sets expectations: what clients may receive between sessions, what remains human, and how consent works. This communicates maturity and helps clients feel safe. It also aligns with ICF guidance on transparency and consent for AI in coaching, making it easier to answer procurement and HR questions without scrambling. Keep both documents short; clarity beats boilerplate every time.

Next, put simple controls around prompts and outputs. Avoid feeding identifiable details to systems without explicit consent. Prefer on-device or enterprise-licensed tools when possible. Keep a light audit trail of the prompts you use for recurring workflows, so you can reproduce good results and spot drift. Finally, consider peer supervision focused on digital practice: discuss use-cases, red-flags and client feedback, just as you would with any other element of the craft. Governance isn’t a bureaucratic exercise; it’s part of how AI in coaching remains ethical, effective and trustworthy as it scales.

Three patterns stand out. First, role-play for difficult conversations. The FT profiled pilots where managers rehearse feedback or conflict scenarios with an AI “coachbot”, practising language and tone before the real thing. Second, always-on companions. CoachHub’s Companion and similar tools sit in chat environments to provide on-demand nudges, reflections and micro-learning. Third, structured guidance for career growth. BetterUp’s AI features support reflection, planning and decision rehearsals, then escalate to humans where depth is needed. In each case, AI in coaching expands access and cadence while humans hold space for complexity. That division of labour is the point, not a flaw.

ICF’s companion “examples” materials make the same distinction: interactive and conversational systems can contribute to behaviour change, but they warrant higher standards for ethics, safety and measurement. Coaches who participate in these ecosystems can influence how they evolve by insisting on clarity, opt-outs and feedback loops. Buyers can do the same by asking providers how they meet the standards and how humans stay in the loop. It’s not enough for AI in coaching to feel useful; it must be demonstrably responsible — and coaches are well-placed to hold that line.

Choose one workflow to improve and make it boringly reliable. For many coaches, that’s post-session follow-through. Use a model to turn your notes into a clean summary and two custom reflection prompts. Send a short recap, agree next steps, and set an automated nudge three days later. Done consistently, this elevates perceived quality without feeling robotic. Over a month you’ll save hours and increase accountability, the sweet spot where AI in coaching shines. Then, once that feels natural, add a discovery-side habit: a monthly insights newsletter or a weekly LinkedIn paragraph derived from anonymised themes you’re seeing in client work.

Keep the human tone by editing every outward-facing piece. Read it aloud; if it sounds like you, publish it. If not, tweak it until it does. Add small touches that only you can add — a case vignette (de-identified), a metaphor your clients recognise, a question you’re sitting with. The aim isn’t volume; it’s rhythm. AI in coaching helps you show up regularly. You supply the judgement, ethics and warmth that turn a workflow into a relationship. Anchor that with transparent consent and you have a system you can be proud of — and one clients will recommend.

Start with the outcomes you already care about and work backwards. If the promise is “better performance conversations,” define what changes for the client, the team and the business. Then decide what signals you’ll track: engagement with prompts, completion of experiments, self-ratings, and manager feedback. Platforms make this easier by aggregating patterns and surfacing trends, but solo coaches can do it with a simple spreadsheet and periodic check-ins. AI in coaching can assist by generating clean dashboards from your data and by spotting themes you might miss when you’re busy. The trick is to keep measures light enough that clients actually use them.

Where organisations are involved, align with their language. HR and L&D leaders care about adoption, retention and performance proxies. Translate your coaching into those terms. If you’re participating in a platform ecosystem, ask how your work will be represented in their analytics and how human coaching is distinguished from automated interactions. If you operate independently, tell clients exactly what you will measure and why — and show them the trendline. When buyers see a clear line from AI in coaching to outcomes, investment follows the evidence rather than the novelty. That’s good for the field and good for your business.

Both. The FT carried a letter from CoachHub’s co-founder arguing that AI can increase personalisation and access for underserved groups by offering immediate insights and exercises. That promise is credible when systems are designed with accessibility in mind: transcripts by default, multiple languages, plain-English options, screen-reader friendly outputs. At its best, AI in coaching can make support available in more places, at more times, to more people. But without attention to bias and context, it can also encode blind spots and reinforce inequity. The solution is design plus governance, not ignoring the tools.

The ICF standards again offer guidance. They encourage teams to test for bias, to invite diverse feedback and to be realistic about what AI can and cannot do. Coaches can adopt the same posture. When you spot bias in outputs, correct it and feed back to vendors. When a client prefers a modality that the tool doesn’t support, adapt. Accessibility is not just a legal box to tick; it is a quality marker for modern practice. Coaches who make inclusion visible — and who use AI in coaching to remove friction for clients — will be chosen over those who treat accessibility as an afterthought.

Pick one container and one cadence. A sensible starting point is the between-sessions layer: recaps, prompts and nudges. Implement a simple stack — your existing notes tool, a writing model for summaries, and your email/CRM to automate timing. Commit to running it with your next five clients, then iterate based on feedback. You’ll find your voice inside the workflow and make it your own. Once that’s steady, add a discovery habit: a monthly insight you publish consistently. Over time, AI in coaching stops feeling like a project and becomes air support for your practice. Small steps, repeated, compound faster than big plans postponed.

If you prefer to learn with others, join a cohort that teaches both the craft and the business. Look for programmemes that integrate ICF competencies, ethical AI use, business development and supervised practice. Steer clear of offers that sell “automated client acquisition” divorced from standards. Your reputation is your asset, and trust is slow to build. Use technology to enhance it, not to gamble with it. Done well, AI in coaching is the quiet assistant that makes you more consistent, not the loud gimmick that erodes your voice.

International Coaching Education (ICE) treats AI as one piece of a larger ecosystem designed to build credible, sustainable practices. Education alone isn’t enough. ICE combines ICF-accredited pathways with strengths development, lifetime community and business acceleration. That means you’re not left to “figure out the business bit” alone; you’re supported to define a niche, practise ethically with peers, and embed low-risk AI in coaching workflows that raise quality without diluting presence. The emphasis is on drip-feed learning, flipped-classroom impact and real practice hours that count towards credentialling.

Because the community is permanent, coaches keep access to reciprocal coaching to build hours, mock exams to demystify assessment, and advanced business development like Roadmap to Revenue — a programmeme that focuses on visibility, confidently closing contracts and designing sustainable offers. Strengths coaching help you articulate a unique value proposition and recognise clients’ talents as levers for change. In short, the model orients around impact and income together: rigorous standards, authentic voice, and disciplined operations — with AI in coaching used to support consistency rather than replace craft.

On 19 August 2025, ICE published our first article in our social media exploring how coaches can use AI to attract their ideal clients and enhance their practice. Since then, we have taken a deep dive into the International Coaching Federation’s (ICF) AI Coaching Framework & Standards (2024), which offer vital guidance on ethical integration of AI. This updated guide weaves together the real-world technologies, market trends, and ICF standards—equipping you with the most timely, credible insight into the future of coaching.

If you read our first version of this article, thank you—we’re glad to now share this expanded edition, aligned with ICF’s new guidance.

The global coaching industry is valued at around US$6.25 billion in 2024, projected to rise to US$7.3 billion by 2025, supported by a robust 17% compounded annual growth rate (CAGR)—highlighting durable client demand. The online coaching market and digital platforms are expanding too: the coaching platform sector is expected to grow from US$3.8 billion (2025) to US$11.1 billion by 2035, a strong 11.2% CAGR. Meanwhile, the wider leadership development market — which includes coaching alongside training programmes and other initiatives — is estimated at over US$100 billion in 2025, with forecasts to nearly double by 2032. Within this, professional coaching is one of the fastest-growing segments, with the International Coaching Federation estimating the coaching industry itself at around US$4.56 billion in 2023, growing steadily year-on-year.

What this means for fees and ROI:
– Entry-level services—driven by AI and platforms—may become more affordable, setting new expectations for access.
– However, human coaches who are ICF-certified, business-savvy, and uniquely positioned are poised to command higher fees and stronger ROI as organisations and individuals seek transformational impact over automated convenience.

Even as AI scales basic coaching, there are elements it simply cannot replicate:
1. Transformational depth: Human coaches bring intuition, emotional attunement, and presence—vital for deep impact.
2. Trust and relational nuance: Coaching is built on trust and complex, empathic relationships that AI cannot replicate ethically.
3. Contextual judgement and ethics: Human coaches interpret culture, organisational complexity, and sensitive dynamics—areas where AI falls short.
4. Professional credibility: Only trained, ICF-accredited coaches uphold the profession’s standards, ethics, and development commitments.

As a result, well-trained coaches with a compelling brand and business acumen will see ROIs rise, while generalised or automated offerings remain commoditised.

Here are notable live examples of how AI is being used in conversational coaching today:

– BetterUp Grow: An AI-augmented coaching platform combining machine learning, behavioural science, and human coaching. Early users report 95% satisfaction and a 16% increase in confidence. Half prefer a hybrid model—AI plus human support.
– CoachHub’s Aimy: An AI coachbot used in workplaces to help managers role-play difficult conversations and develop leadership skills affordably. Organisations like HubSpot have adopted it as a scalable support tool.
– Valence’s Nadia: A multilingual AI coach deployed widely at WPP to support managerial development with confidential, 24/7 role‑play tools.
– Academic insight—LLM-based chatbot in blended coaching: Research shows that combining LLM-powered chatbots with human coaches enhances self-reflection and leadership development—highlighting AI’s supportive, not replacement, role.

ICF identifies four application types for AI in coaching. These vary in risk and purpose:

AI is reshaping the coaching market—expanding access, streamlining workflows, and setting sharper expectations. Market growth is rapid, especially in digital and online coaching segments, where structured and scalable solutions are increasingly supported by AI. Yet, human coaches remain indispensable for deep trust, ethical guidance, and transformational impact.

The key is intentional adoption. Let AI handle the repetitive, the structured, and the scalable, while keeping humans at the centre for authenticity, ethics, and meaningful change. Following ICF’s guidance ensures adoption is transparent, consent-based, and bias-aware. Coaches who align with strong professional standards, clear positioning, and ethical integration of AI won’t be replaced by technology—they’ll be differentiated by how responsibly and skillfully they use it.

At ICE, as the only ICF-accredited provider combining coaching education, credentialing support, structured Business Acceleration, Strengths Coaching, lifetime community, and ongoing learning with custom pacing, we prepare coaches not only to adapt but to thrive. By blending professional credibility with business acumen and thoughtful AI integration, our coaches lead the high-end, high-impact segment of the market—one where technology supports discovery and delivery, but never overshadows the human transformation at its core.

Taymour Miri is an ICF master coach and a Gallup certified strengths coach and more recently one of the first 136 coaches world wide to be awarded an Advanced Certificate in Team Coaching. He has 30 years’ experience in leadership roles and 20 years of experince in coaching. Taymour has trained over 1,500 coaches across five continents and is the founder of International Coaching Education (ICE).