Your board member read the AI story on the way to your last meeting. By the time they sat down, the question on their list was why your competitors seem to be moving faster than you.
That is not a bad-faith question. Boards are doing exactly what boards should do. Their job is to ask it. Your job, as the founder, is to answer in a way that reads as authority, not apology. Cave and you ship activity designed to look like results. Stonewall and you look complacent. There is a third move, and it starts with understanding what a realistic timeline actually is.
What does a realistic AI timeline actually look like?
Meaningful AI returns typically take twelve to twenty-four months to appear on the numbers. MIT research published in 2025 found around 95% of generative AI pilots produce no rapid revenue acceleration, with failure traced to workflow integration rather than model quality. BCG’s analysis puts roughly half of organisations stuck at proof-of-concept, unable to scale further.
The reason is mostly structural. Effective AI implementation is a change-management exercise as much as a technology project. Research on technology adoption consistently shows that tools fail not on technical merits but when the people and leadership work around them is underestimated. Your team needs time to genuinely change how they work, not just which tools they use. That shift rarely completes in a single quarter.
Kyndryl’s 2024 research adds a further dimension. Around 70% of leaders say their workforce is not ready for AI, and only 14% report having aligned their workforce, technology and growth goals. That readiness gap is where the time actually goes. The model is often the straightforward part. Changing how a team works around it is where twelve to twenty-four months come from.
The timeline is a forecast, not an apology for slow progress. A board that understands this can hold its nerve and read the work you are doing correctly. A board working from headline expectations cannot, and the founder who lets headline expectations set the agenda will find the gap between board perception and operational reality increasingly hard to manage.
Why does the gap between board expectation and operational reality matter?
Cave to board pressure and you end up running theatre, visible AI activity designed to look like progress rather than produce it. Kyndryl’s 2024 research found only 14% of organisations have genuinely aligned their workforce, technology and growth goals around AI. Running pilots before that groundwork exists means the numbers will eventually reveal what the board wanted to avoid seeing.
Caving does not only produce a bad result. It sets a precedent. When you give ground because the board applied pressure, you teach the board that pressure works. The founder who sets an expectation and then misses it is in a measurably weaker position for every board conversation that follows.
Stonewalling is not the answer either. A board that reads your position as complacency rather than strategy has a different concern, one that the numbers alone will not address. The goal is not to resist but to explain with enough specificity that resistance becomes unnecessary. Most of the tension in these conversations comes not from disagreement about the goal but from a mismatch in what each side believes the timeline should look like.
Where does your board’s impatience with your AI timeline come from?
Board members are not inside the operation. Their picture of AI progress comes almost entirely from the same headlines, conferences, and peer conversations you encounter yourself. The NACD’s 2025 survey found 62% of director respondents now set aside dedicated board agenda time for AI, up from 35% in 2024. That pace of board attention creates its own pressure on the questions they bring to the table.
There is also a governance dimension. In investor-backed businesses, boards are accountable to their own stakeholders for whether the companies they oversee are keeping pace. Spencer Stuart’s research on the CEO role in AI found boards are increasingly concerned when founders have delegated AI entirely without maintaining personal fluency. From the board’s position, “we’re working on it” without a clear framework can read as being behind rather than being deliberate.
BCG’s 2026 research is worth naming directly. Around 60% of CEOs report feeling their boards are rushing AI decisions. The research also notes that board members with lower personal confidence in AI tend to push hardest. The people least equipped to judge a realistic timeline are often the most impatient with one. Understanding that asymmetry before you walk into the room changes how you prepare.
The founder who understands where the pressure comes from is better positioned to address the actual concern rather than the surface question. Your board member has a legitimate concern about AI progress. They are simply working from incomplete information about what it looks like in practice. That is a communication gap you can close.
When should you hold firm and when should you give ground?
Defending a timeline and apologising for one are not the same move. The founder who defends names the sequence, their current position in it, and the specific cost of accelerating too soon. The founder who apologises says it is taking longer than expected. One signals authority. The other signals the timeline is negotiable, and it will be followed by another push.
There is a third option, and it is the one that resolves the tension. You give ground on the visibility of the process, not on the timeline itself. You offer the board something concrete to follow across quarters rather than a date they can hold you to.
That means being specific about what board-ready looks like in this phase. Quarterly updates with a consistent shape, covering what was done, what changed, and what comes next. Not a summary of activity. A story the board can follow as it builds, so they are reading a coherent narrative over four quarters rather than disconnected reports. BCG’s 2026 research on CEO leadership in AI is explicit that boards now expect the conversation to shift from predictions to process, showing how the work connects to specific business improvements rather than making forward claims about financial impact. That framing works in your favour.
What do you put on the table instead of a date?
The compounding-value argument is the most effective thing a founder can place in front of a board. The work that takes the most time and produces the least visible output in the short term, getting the team genuinely fluent, making data accessible, documenting the workflows, is also what makes everything that follows faster. Boards that understand this measure trajectory rather than the calendar. Boards that do not will keep asking for a date.
The progress markers that work are operational rather than financial. Your operator can now run a process that previously needed a founder sign-off. A team has completed training that changes how they handle a specific workflow. Data that previously lived in someone’s head is now documented and accessible to the right people. Each of these is verifiable. None requires a revenue line to prove it happened.
When you present these markers consistently, over three or four quarters, something shifts in the board conversation. They stop asking when the return will come and start asking which marker comes next. That is a different dynamic entirely, and it is the one a realistic timeline makes possible.
The founder who frames twelve months of readiness work as twelve months of foundation building is doing something more than managing the board. They are also protecting the business from the pressure to produce visible results before the valuable ones are ready. That sequencing is the argument, and it is what the work eventually lands on.
If you want help preparing a board conversation that holds that ground with specificity, book a conversation.
