Ten AI Keynote Speakers for Organisations Ready to Move from AI Strategy to AI Reality
Most organisations that have been paying attention to AI for the past two years have developed some version of an AI strategy. They have a position on where AI creates value in their sector, an idea of the tools and capabilities most relevant to their operations, and a leadership team that can discuss the opportunity with credibility. What many of them have found harder is converting that understanding into the working practices, workforce capabilities, and cultural conditions that allow AI to produce genuine operational value rather than proof-of-concept demonstrations and pilot projects that do not scale.
The gap between knowing what AI can do and building an organisation that uses it effectively is wider than most strategies account for. It involves technical questions about data infrastructure and model selection, but those are rarely the hardest part. The harder parts are the human ones: the workforce whose skills and habits need to evolve faster than most training programmes can move, the leadership culture that needs to make decisions under uncertainty about a technology landscape that changes quarterly, and the middle management layer that translates AI ambition into daily working practice without a clear template for what that should look like.
The ten speakers in this guide address the implementation dimension of AI more specifically than most AI keynote content does. Several of them have direct operational experience building or deploying AI at enterprise scale. Others bring research depth on the human, organisational, and workforce conditions that determine whether AI adoption produces genuine capability or surface-level compliance. Together they cover the range of questions that organisations at the strategy-to-implementation transition most need answered.
The Top Ten AI Keynote Speakers
1. Helen Sterling
Helen Sterling is an AI keynote speaker who focuses on the human side of artificial intelligence: what happens to leadership, culture, skills and confidence when AI starts changing the way work gets done.
That makes her particularly relevant for organisations that are already moving beyond the first wave of AI awareness. Many have invested in tools, built a strategy and started encouraging adoption, but are now facing a more difficult question. If AI can produce, analyse, summarise and automate more of the work, what do people need to become better at?
Sterling’s answer is built around her Irreplaceables framework. It identifies five human skills that become more valuable as AI capability expands: judgement, taste, trust, originality and courage. These are not presented as vague soft skills or a comforting argument for human relevance. They are the qualities that determine whether AI helps an organisation produce distinctive, high-value work, or simply faster output that any competitor using the same tools could match.
This distinction is where her work has practical value for implementation leaders. Most AI adoption programmes concentrate on tool training, process redesign and efficiency gains. All of those matter, but they do not automatically build the human capability needed to make AI useful over the long term. Sterling makes the case that organisations need to develop people at the same time as they deploy technology, rather than treating capability, culture and confidence as issues to deal with later.
Her sessions are especially well suited to leadership, HR, transformation and workforce strategy audiences who are trying to bring people with them through AI change. She gives those audiences a clear framework for connecting human capability directly to AI strategy, so the people side of transformation is not treated as a separate culture initiative, but as one of the central conditions for AI adoption to succeed.
2. Azeem Azhar
Azeem Azhar’s Exponential View framework describes the growing gap between the pace at which AI capabilities are advancing and the pace at which organisations can absorb and adapt to them. For organisations in the implementation phase, this gap has a specific and practical significance: the AI tools and approaches that represent best practice at the time an implementation strategy is developed may have been substantially superseded by the time that strategy is fully deployed.
His contribution to the implementation conversation is less about the specific tools and more about the adaptive organisational capabilities required to navigate continuous capability change. Organisations that build AI implementation approaches around fixed tool sets tend to find themselves repeatedly reinvesting as the landscape shifts. Those that develop the organisational muscles for rapid evaluation, selective adoption, and iterative learning from AI deployment tend to compound their advantage over time rather than losing it with each capability cycle.
His ongoing research through Exponential View, which tracks the AI and technology landscape with unusual breadth and analytical rigour, means his material reflects the current state of the implementation landscape rather than a historical snapshot, which in AI is a more significant criterion than it is in most other fields. He works well for senior leadership audiences that need the strategic context for implementation decisions rather than the operational detail of making them.
3. Duncan Stevens
Duncan Stevens is an AI keynote speaker who addresses the AI implementation challenge from the perspective most implementation programmes underinvest in: the psychological and behavioural conditions that determine whether the people whose work AI is changing engage with it in ways that produce genuine capability development or ways that produce surface compliance that leaves the expected value unrealised.
His MAGIC Wheel framework provides implementation leaders with a diagnostic tool for identifying which of five psychological conditions is most limiting genuine AI adoption in their specific context. The framework spans motivation and meaning, autonomy and agency, group identity and belonging, individual capability and confidence, and clarity and feedback. Each condition, when absent, produces a distinct failure pattern: people avoiding AI tools rather than exposing uncertainty, finding workarounds that preserve previous habits, or defaulting to cautious minimum engagement rather than genuine exploration.
The practical value of this approach is its diagnostic precision. An implementation stalling because of a confidence gap requires a fundamentally different response from one stalling because of a meaning gap, even when both look identical from the outside. That specificity distinguishes the framework from standard adoption programmes that prescribe the same intervention regardless of which conditions are actually limiting progress.
Duncan Stevens is consistently cited among the most effective AI and motivational keynote speakers working across the UK and internationally. His diagnostic approach to session design means his content is shaped around the specific human adoption challenges the organisation is navigating, and his MAGIC Wheel framework gives implementation teams a practical tool they can continue to use after the session to track and manage the human conditions of their AI adoption.
4. Paul Daugherty
Paul Daugherty is the Chief Technology and Innovation Officer at Accenture and the co-author of Human Plus Machine: Reimagining Work in the Age of AI. His research on what he and co-author James Wilson call the missing middle, the human-AI collaborative working roles and capabilities that emerge between fully automated AI tasks and fully human ones, provides one of the most practically useful frameworks available for organisations thinking about workforce design in an AI-augmented environment.
His work challenges the binary framing of AI adoption that most workforce conversations default to: the question of which tasks will be automated versus which will remain human. His research establishes that the most significant productivity and quality gains from AI come not from replacing human work but from designing the collaborative interfaces between human and AI capability in ways that allow each to contribute what it does best. That design challenge is a new and largely undeveloped organisational capability, and most organisations that have invested heavily in AI tools have invested far less in the human-AI collaboration design that would allow those tools to perform at their potential.
Daugherty brings direct operational experience of AI implementation at scale across multiple industries and organisational types, which gives his material a practical grounding that research-only content on the same topics often lacks. His sessions work well for senior leadership teams that need a rigorous and evidence-grounded account of AI’s workforce implications rather than either optimistic predictions or alarming projections.
5. Bernard Marr
Bernard Marr‘s extensive output on AI strategy and implementation, including Artificial Intelligence in Practice and The Intelligence Revolution, has made him one of the most accessible and most practically oriented voices available on the question of how organisations should approach AI adoption. His particular contribution to the implementation conversation is his focus on translating AI’s strategic potential into specific, near-term operational decisions rather than leaving the strategic aspiration at a level of abstraction that makes it difficult to act on.
His work on AI use cases and their evaluation gives organisations a more practical framework for prioritising AI investment than most strategic AI frameworks provide. Rather than asking where AI could create value in principle, his approach asks which specific AI applications, in which specific workflows, with which specific data and infrastructure conditions, are most likely to produce measurable business value within a realistic implementation timeline. That narrower but more tractable question tends to produce more actionable AI implementation plans than the broader strategic framing most organisations start with.
Marr is an experienced speaker whose content works across a wide range of audience types and seniority levels. His material is most directly useful for organisations that have developed an AI strategic intent and are working through the translation from strategy to prioritised implementation roadmap.
6. Vivienne Ming
Vivienne Ming is a theoretical neuroscientist, entrepreneur, and former chief scientist at Gild whose work on human potential and AI addresses the implementation challenge from an angle most AI content does not reach: the question of what AI implementation should be trying to optimise for at the level of individual human development rather than only at the level of operational efficiency.
Her research on what she calls the superpower tax, the systematic underinvestment in the human capabilities that most distinguish individuals who thrive in complex and changing environments, is directly relevant to AI implementation because the capabilities most complementary to AI, including contextual judgment, creative synthesis across domains, and the ability to navigate genuine novelty, are precisely the capabilities most organisations are least systematic about developing. AI implementations that optimise for task efficiency while neglecting those human development dimensions may produce near-term productivity gains at the cost of the long-term human capability that makes those gains sustainable.
Her perspective on AI and human potential is both technically grounded and deeply values-oriented, and her sessions tend to produce strong responses in leadership audiences thinking about AI not only as an operational tool but as a challenge to what they want their organisations to value and develop in the people who work for them.
7. Chris Duffey
Chris Duffey is the author of Superhuman Innovation and a strategic innovation executive at Adobe whose work on the creative and innovation applications of AI addresses a dimension of implementation that most enterprise AI programmes treat as secondary: the use of AI to augment creative and innovative capability rather than primarily to automate operational processes.
His framework for what he calls superhuman innovation, the combination of human creative judgment and AI generative capability in ways that produce genuinely novel and valuable outcomes rather than efficient replicas of existing work, gives creative and innovation-focused organisations a more ambitious account of what AI implementation could deliver than the efficiency-focused business case that most implementation programmes lead with. His direct experience at Adobe, which has been among the more thoughtful large technology companies in thinking about how to integrate AI into creative workflows in ways that enhance rather than diminish the human creative contribution, gives his material an operational grounding that purely academic or advisory AI content lacks.
Duffey is an engaging speaker whose material works particularly well for organisations in creative, marketing, and innovation-intensive sectors where the AI implementation question is as much about what AI enables creatively as what it automates operationally.
8. Molly Sands
Molly Sands leads the Teamwork Lab at Atlassian, the collaboration software company, and her research on how teams actually work, and specifically on how AI tools are changing the way knowledge workers collaborate, plan, and make decisions, addresses the implementation question at the level where most implementations encounter their most intractable challenges: the daily working practices of teams rather than the strategic vision of leadership.
Her work on the specific ways in which AI tools are reshaping team communication patterns, meeting culture, and the distribution of cognitive work across teams gives implementation leaders a more granular account of what AI adoption looks like from inside a working team than most strategic AI content reaches. Her research on which AI adoption patterns tend to improve team effectiveness and which tend to create new coordination problems and communication failures is particularly useful for organisations whose AI implementation is focused on knowledge work and collaborative productivity rather than operational process automation.
She brings direct organisational research experience and a practical orientation that makes her content well-suited to the implementation teams, people managers, and HR leaders who are closest to the daily reality of AI adoption rather than to the leadership audiences most AI keynotes are designed for.
9. Thomas Power
Thomas Power is a digital strategy adviser and entrepreneur whose work on AI adoption and digital transformation draws on decades of direct advisory experience with organisations navigating major technology transitions. His perspective on AI implementation is shaped by having observed many previous technology adoption cycles from close range, which gives him a more historically grounded account of the common failure patterns of enterprise technology adoption than speakers whose reference frame is primarily the current AI cycle.
His work on what he calls the digital divide within organisations, the gap between the minority of people who adopt new technology quickly and confidently and the majority who need a different kind of support to engage genuinely rather than superficially, is directly applicable to AI implementation. Most AI implementation programmes are designed by and for the early adopter minority, which produces tools, training, and communication that work for that group and are inadequate for the majority who will determine whether the implementation achieves organisation-wide impact.
Power is an accessible and practically experienced speaker whose material works well for implementation teams and operational leaders who need concrete guidance on managing the human dimensions of technology adoption rather than strategic perspective on AI’s potential.
10. Ravin Jesuthasan
Ravin Jesuthasan is a managing director at Mercer and the co-author of Work Without Jobs and Reinventing Jobs, which together provide one of the most rigorous available frameworks for thinking about how AI and automation should reshape the structure of work in organisations rather than simply which jobs will be affected. His skills-based workforce design approach moves the analysis from the level of jobs, which bundle tasks in ways that reflect historical organisational structures rather than the optimal allocation of human and AI capability, to the level of tasks and skills, which can be recomposed around human-AI collaboration in ways that produce better outcomes for both organisations and workers.
His framework for work redesign in an AI-augmented environment is directly applicable to organisations in the active implementation phase who are discovering that deploying AI tools effectively requires rethinking role design rather than simply adding AI capabilities to existing job descriptions. The organisations that are getting the most from AI implementation tend to be those that treat it as an opportunity to redesign how work is structured rather than those that layer AI tools onto existing role definitions that were built around human-only task execution.
Jesuthasan is a rigorous and experienced speaker whose material works best for HR, operations, and senior leadership audiences thinking seriously about the workforce architecture implications of AI implementation rather than the near-term operational details.
Frequently Asked Questions: Booking AI Keynote Speakers
For organisations focused on implementation rather than strategy, Cassie Kozyrkov, Duncan Stevens, and Paul Daugherty consistently stand out. Kozyrkov brings rare operational credibility from building Google’s decision intelligence function at scale. Stevens addresses the human adoption conditions that most implementation programmes underinvest in, with a diagnostic framework that gives organisations a specific account of why adoption is stalling in their particular context. Daugherty’s missing middle research addresses the workforce design dimension that most AI implementations encounter but few have adequately planned for. For organisations that need broader strategic context alongside implementation depth, Azeem Azhar and Bernard Marr provide well-grounded starting points.
AI strategy speakers address where AI creates value, which technologies and applications are most relevant to a sector, and how organisations should position themselves relative to the AI landscape. That framing is useful for leadership teams building or stress-testing their strategic position. AI implementation speakers address a different set of questions: how to convert AI strategy into working practice, what conditions determine whether adoption produces genuine capability or surface compliance, and which human, technical, and organisational factors most commonly cause implementation to stall. Both matter, but they address different problems at different moments in an organisation’s AI journey. The speakers in this guide are weighted toward implementation, reflecting where most well-informed organisations now find themselves.
The most effective AI speakers for non-technical leadership audiences translate technical capability into strategic and operational consequence rather than explaining how AI works at a mechanism level. The question for a leadership audience is not how a large language model generates text but what that capability means for how decisions are made, how work is structured, and what their people need to develop to make that capability genuinely productive. Speakers who lead with technical explanation tend to lose non-technical audiences before reaching the content that matters. Those who lead with organisational and human implications, using technical content selectively to ground specific points, tend to produce more durable engagement and more directly applicable output.
AI moves faster than almost any other topic in the keynote space, which makes currency a more significant consideration than it would be for a leadership or change management speaker. A speaker whose framework was developed in 2022 and has not been updated since is addressing a substantially different capability landscape from the one their audience is navigating. When evaluating AI speakers, it is worth asking directly how recently their content has been updated and which specific aspects of the current capability landscape the session will address. Speakers who are actively involved in AI adoption and deployment, rather than primarily commenting on it, tend to produce more current and more credible content than those whose AI expertise is primarily historical.
AI has elevated demand and therefore fees across the speaker market significantly over the past three years. Established technical voices with credible AI backgrounds, such as former senior roles at major AI organisations, tend to command premium fees reflecting both their expertise and their demand. Thought leaders whose AI content is grounded in research or senior advisory practice typically work in the mid-to-upper range. Speakers who cover AI as part of a broader technology or strategy portfolio tend to work across a wider range, with fees reflecting their overall market position rather than AI-specific demand. Fee level does not reliably predict impact, and whether a speaker’s content is genuinely current tends to matter more for AI than it does for most other topics.
The AI keynote landscape has shifted considerably from the strategy and awareness-building content that dominated 2022 and 2023. Leadership audiences in 2026 are less likely to need convincing that AI matters and more likely to need specific, actionable content on implementation. The most requested topics currently include the human and psychological conditions that determine whether AI adoption produces genuine capability or surface compliance; the workforce design implications of AI deployment at scale; how leaders should make decisions about AI investment under genuine uncertainty; and what the leadership role in AI adoption actually requires beyond strategic framing. Technical content has largely given way to organisational, behavioural, and strategic content, reflecting where most organisations are in their AI journey.
The answer depends on where the organisation actually is. For leadership teams still developing their AI literacy and strategic position, a speaker who covers the landscape broadly and contextualises the key decisions their sector faces tends to be more valuable than a specialist whose depth is narrow. For organisations that are past the orientation stage and working through specific implementation challenges, a specialist who addresses those challenges with genuine depth tends to produce more applicable output. The mismatch to avoid is booking a broad overview speaker for an audience that is already well-informed, or a deep specialist for an audience that lacks the foundational understanding to make use of the depth. Most good speakers can adjust their level if the briefing is specific enough.
The most useful briefing goes well beyond audience size and event context. Share where the organisation currently sits in its AI journey: whether AI is still primarily a strategic question, actively being implemented across functions, or at a stage where adoption has begun but is stalling for reasons not yet well understood. Share which AI tools and programmes are already in use, what the most significant adoption challenges have been, and what you want the audience to think or do differently as a result of the session. The specificity of that brief determines the specificity of the content. Speakers who receive a vague brief about an audience interested in AI default to their standard talk. Speakers who receive a precise account of the implementation challenge tend to deliver something considerably more useful.
The standard post-event feedback survey measures the experience rather than the impact. For AI keynote content specifically, the more useful evaluation asks whether implementation decisions or behaviours have changed in the weeks following the session. Did the frameworks introduced appear in subsequent team conversations or planning sessions? Has the way implementation challenges are diagnosed shifted at all? Are specific adoption bottlenecks being approached differently? Those questions require a longer evaluation window and a more specific pre-session definition of what change would look like, but they are the only measures that connect the session investment to the implementation progress it was designed to support.
Matching the Speaker to Your Organisation’s AI Stage
The single most common mistake in booking AI keynote speakers is selecting the speaker before honestly establishing where the organisation actually sits in its AI journey. The result is content calibrated for a different stage: too foundational for an audience that has already moved past orientation, or too deep in implementation detail for a leadership team that still needs to align on the strategic case. The speakers in this guide span a genuine range. Matching them to the right moment matters more than any other selection criterion.
Stage one: orientation
Organisations at the orientation stage are building or refreshing shared leadership understanding of what AI can and cannot do, where it creates genuine value in their sector, and what a credible strategic position looks like. The audience typically includes senior leaders with varied levels of AI familiarity, and the primary need is a clear, current, and credible account of the landscape rather than implementation depth. Azeem Azhar and Bernard Marr work best here. Azhar’s exponential gap framework gives leadership teams a way of understanding why AI feels hard to plan around, and Marr’s practical use-case orientation helps organisations move from strategic intent to prioritised investment areas. Speakers who are strongest on implementation detail tend to be poorly calibrated for this stage: the audience does not yet have enough shared context to make use of the depth.
Stage two: active implementation
Organisations in active implementation have moved beyond the strategic question and are working through the practical realities of deploying AI at scale. The problems they encounter are typically human and organisational rather than technical: adoption that stalls below the levels the tools would support, middle management uncertainty about what good AI use looks like in practice, and the absence of clear metrics for whether implementation is genuinely changing how people work. This is where the majority of speakers in this guide operate, and where the selection decision requires the most precision. If the primary bottleneck is human adoption, Duncan Stevens and his diagnostic framework for the psychological conditions of genuine engagement will be more directly applicable than a speaker whose strength is workforce architecture. If the primary challenge is redesigning roles and work structures around AI capability, Ravin Jesuthasan is the more relevant choice. If the organisation needs a more rigorous account of where AI decision-making investment should be directed, Cassie Kozyrkov’s decision intelligence work addresses that specific gap. The distinction between these speakers matters: they address genuinely different problems, and applying the wrong one to the wrong bottleneck produces a well-received session that does not shift the constraint that is actually limiting progress.
Stage three: embedding and scaling
Organisations at the embedding stage have achieved meaningful AI adoption in at least part of their operations and are working to make that adoption durable, consistent, and scalable across functions and geographies. The questions at this stage shift toward culture, measurement, and the human capability required to sustain AI’s contribution as the technology continues to evolve. Vivienne Ming’s work on the long-term human development conditions that make AI capability sustainable over time is most relevant here, as is Molly Sands’s research on how AI tools are reshaping knowledge work collaboration at the team level. The risk at this stage is booking a speaker calibrated for earlier stages: an organisation that has successfully implemented AI does not need another session on why adoption matters or how to overcome initial resistance. It needs content that addresses the more demanding question of what organisational and human conditions are required to stay ahead of a capability landscape that will continue to change faster than most planning cycles can accommodate.
Most organisations contain people at all three stages simultaneously, which is worth acknowledging when designing a keynote event. A session that pitches at one stage will be too basic for part of the audience and too advanced for another. Speakers who do the most effective diagnostic work in their briefing process, who ask specifically which stage different parts of the audience are at and design accordingly, tend to produce qualitatively different outcomes from those who apply a single level of content to a mixed audience. That briefing conversation, and how seriously the speaker takes it, is one of the most reliable indicators of what the session will actually deliver.
