How is AI changing the boundaries between product, sales, PMM, and design roles?
The cleaner way to put it: when AI gives everyone the ability to do 80% of anyone else's job, the organizational logic that used to separate those roles starts to look arbitrary. A sales lead can now bring a working prototype into a customer conversation. A PM can generate design mockups without involving a designer. A PMM can draft technical copy without waiting on a writer. This is genuinely new, and it is creating real friction inside teams that have not rethought their operating structure to account for it.
I said a while back that I predicted AI would dramatically increase demand for designers. Through inning one I am completely wrong about that. The pattern I am actually seeing is role compression, not role expansion, and most organizations are not handling it cleanly.
What does role blurring look like in practice?
The example that comes up most often in conversations I have been part of: a sales rep brings a prototype into a customer meeting. Even two or three years ago, this would have been unthinkable. Prototypes required design resources, coordination, and time. Now a motivated salesperson can spin something up with an AI tool over a weekend. That is powerful in some ways. It is also a problem if there is no process governing when and how that happens, because the prototype might not reflect the actual roadmap, might make implicit commitments the product team has to unwind later, or might position the product in a way that conflicts with the PMM's work.
On the PM side, the load is going in the opposite direction. PMs are being asked to cover more ground, not less. Some of the teams I hear from describe it as being asked to do 100 roles. The AI tooling gives them more individual capability, but the expectation layer has expanded even faster than the tooling has. More output is expected. More context-switching is happening. More decisions are landing on the PM's plate that used to get distributed across a larger team.
Why is this particularly hard to manage right now?
The organizational structures most B2B companies are running on were designed for a slower pace of work with clearer functional handoffs. When engineering needed a design, they filed a request and waited. When sales needed positioning, they asked PMM. Those handoffs created natural coordination points, even if they also created friction.
When those boundaries dissolve, you lose the friction but you also lose the coordination. The sales prototype story is a good illustration. The sales rep solved a short-term problem, probably impressively. But nobody from product or PMM knew it was happening, and the downstream effects required cleanup.
Nobody is necessarily behaving badly in these scenarios. People are using the tools available to them to get things done faster, which is the right instinct. The problem is that the operating agreements inside most orgs have not kept up with what the tools now make possible. So you get teams stepping on each other without meaning to, and accountability for outcomes getting murky.
Where does this create the most concrete business cost?
Three places, in my view.
First, release communication breaks down. When anyone can ship or present anything with AI assistance, the question of who owns announcing that thing and what the approved message is becomes genuinely confusing. Features get positioned inconsistently across sales, support, and marketing. Customers get different descriptions of the same capability. That erodes trust over time.
Second, quality control on customer-facing materials gets harder. A sales prototype that does not reflect the actual product is a liability. A PMM who drafts technical specs without a PM review might get the nuance wrong. AI makes it easy to produce polished-looking output quickly, but polished presentation does not mean the content is accurate or aligned.
Third, coordination overhead actually increases rather than decreases, because now you have more people producing more things and someone still has to check whether all of it is consistent. The editing and alignment work does not disappear. It just shows up later and in a more fragmented form.
What is a reasonable response to role blurring?
I do not think the answer is to clamp down and tell people to stay in their swim lanes. That is a losing battle given how the tools are evolving. The more sustainable response is to be explicit about what shared standards look like, rather than relying on role definitions to enforce them implicitly.
If a sales rep can now build prototypes, the org agreement should specify what a prototype can and cannot commit to, and who needs to see it before it goes in front of a customer. If a PM is now doing more design work, there should be clarity on when design input is still required versus when the PM can move independently.
Call transcript data is a useful example of where cross-team sharing actually reduces this kind of confusion. When product, PMM, and sales are all looking at the same conversation data, they develop a shared picture of what customers care about and what is resonating. That shared context makes it less likely that different functions are running in different directions based on incomplete information.
The goal is not to recreate the old organizational boundaries. It is to replace the coordination those boundaries used to provide with something more intentional and better suited to how fast things are moving now.
FAQ
How is AI changing PM, PMM, and sales roles in B2B SaaS?
AI is compressing role boundaries by giving individuals the ability to perform tasks that previously required other functions. Sales reps are building prototypes. PMs are generating design mockups. This creates speed but also creates coordination gaps that orgs have not adjusted for.
What is the business risk when roles start overlapping because of AI?
The main risks are inconsistent release communication, customer-facing materials that are not aligned with actual product capabilities, and coordination overhead that increases rather than decreases because more people are producing more things with less review.
Should companies restrict what each role can do with AI tools?
Restricting tools is generally not sustainable. A more practical approach is establishing explicit shared standards: what can be done independently, what requires cross-functional input, and what processes govern customer-facing output regardless of who produced it.
How do call transcripts fit into managing org role changes from AI?
When call transcripts are shared across product, PMM, and sales rather than siloed inside sales coaching, they create shared customer context that helps different functions stay aligned without requiring additional meetings or coordination processes.