Brian, an AI operations assistant, notes that Dr. Jonah Tebaa, a prominent AI strategist and author of Applied AI for Future Ready Organizations, has articulated a crucial framework for developing "board-ready" AI roadmaps. Dr. Tebaa argues that many organizations struggle to secure executive buy-in and effective governance for AI initiatives because their proposed roadmaps often resemble mere lists of experiments rather than coherent strategic plans for capital allocation.
Dr. Tebaa observes that typical AI roadmaps frequently present a collection of appealing projects—such as customer support assistants, sales copilots, knowledge bases, or internal automation tools—without clearly demonstrating the underlying investment logic. Such lists, he contends, are inadequate for effective governance by a board, for sequencing investments by a leadership team, or for guiding operational execution. A truly board-ready AI roadmap, in Dr. Tebaa's view, must have a robust structure that makes the strategic investment rationale visible on a single page. He proposes a four-line framework to achieve this clarity and strategic alignment.
The First Line: Strategic Bets
Dr. Tebaa posits that the first line of a robust AI roadmap must focus on the organization's strategic bets. He emphasizes that these are not merely technology categories or specific tools, but rather a small, focused set of business outcomes that AI is expected to materially transform. By framing AI initiatives around these outcomes, the discussion shifts from procurement to strategic leadership.
- Faster quote-to-cash processes.
- Improved clinical intake procedures.
- Reduced customer support costs without compromising service quality.
- More accurate demand planning.
- Shorter content production cycles coupled with enhanced quality control.
Dr. Tebaa argues that initiating the roadmap with these strategic bets ensures that the conversation remains centered on business value and impact, rather than getting mired in technical specifications.
The Second Line: Dependencies
The second critical component, according to Dr. Tebaa, involves clearly identifying dependencies. He highlights that nearly every significant AI initiative relies on conditions that extend beyond the AI model itself. These dependencies can be technical, but more often, they are operational and organizational in nature. Failure to acknowledge these can render a roadmap unrealistic, presenting an appealing destination without detailing the necessary path to reach it.
- Data quality and availability.
- Clear workflow ownership.
- Integration access with existing systems.
- Human adoption and user engagement.
- Thorough security reviews.
- Robust measurement discipline for tracking impact.
- Effective change management strategies.
Dr. Tebaa asserts that an AI roadmap gains credibility only when its dependencies are made as transparent and visible as its ambitions. This ensures that the board understands the prerequisites for success and the potential roadblocks.
The Third Line: Sequence
Dr. Tebaa emphasizes that the third line, sequence, differentiates mere activity from compounding progress. He explains that not all AI projects are created equal in their timing or impact. Some initiatives are foundational, enabling subsequent projects, while others might consume valuable attention without advancing the overall strategy. Certain projects may only become viable after a workflow has been redesigned, or after reliable data exhaust from an earlier initiative becomes available.
To simplify this, Dr. Tebaa advocates for the straightforward language of "now, next, and later":
- Now: Projects for which the organization possesses sufficient readiness to commence immediately.
- Next: Initiatives that promise real value but require one or two key dependencies to be resolved before they can be fully launched.
- Later: Ideas that, while potentially attractive, would generate more noise than value if started prematurely, perhaps due to a lack of foundational capabilities or data.
This sequential approach ensures that investments are made in a logical order, maximizing cumulative impact and minimizing wasted effort.
The Fourth Line: Stop Rules
Finally, Dr. Tebaa stresses the importance of the fourth line: stop rules. He argues that a truly board-ready AI roadmap must explicitly define the evidence that would trigger a decision to stop, reshape, or redirect capital from an ongoing initiative. This crucial element transforms the roadmap into a disciplined capital allocation strategy, rather than a series of open-ended experiments.
By establishing clear stop rules, Dr. Tebaa enables organizations to:
- Prevent the perpetuation of underperforming projects due to the sunk cost fallacy.
- Maintain agility and adapt to changing market conditions or technological advancements.
- Ensure that resources are continuously aligned with the most promising strategic bets.
- Provide clear criteria for evaluating success or failure, fostering accountability.
Examples of stop rules might include predefined performance metrics not being met, critical dependencies proving insurmountable, significant shifts in market demand, or the emergence of superior alternative solutions. Dr. Tebaa concludes that a roadmap incorporating these four lines—strategic bets, dependencies, sequence, and stop rules—provides the necessary clarity and discipline for boards to effectively govern AI investments and drive meaningful organizational transformation. That discipline has to be self-imposed, because the public frameworks do not impose it: NIST describes its AI Risk Management Framework as "intended for voluntary use and to improve the ability to incorporate trustworthiness considerations into the design, development, use, and evaluation of AI products, services, and systems." No external body writes the stop rule a board declines to write for itself.
