When Canada’s first Minister of Artificial Intelligence, Evan Solomon, took the stage at ALL IN 2025 Conference, one idea from him stood out to me: trust is the accelerator of adoption of an AI product.
When people feel safe with AI, the speed at which society embraces new products increases dramatically. That insight resonated because, for product marketers, adoption is everything.
In the race to integrate artificial intelligence into go‑to‑market (GTM) strategies, the promise of speed is seductive: faster content generation, instant insights from a sea of data, hyper‑personalized messaging, and campaign. Yet each capability introduces new risks around ethics and brand reputation.
AI hallucinations, over‑promised capabilities, and “AI‑washed” marketing can quickly erode the credibility that growth depends on: trust. Harvard Business Review puts it plainly: AI has a trust problem.
McKinsey & Company emphasizes that explainability (the ability to articulate what went in as input for the AI model and why a model produced a given recommendation) is central to building confidence among customers and regulators.
Bain & Company recent research on consumer attitudes toward agentic AI shows that while experimentation with AI commerce tools and processes is high among customers, comfort with AI making purchases on their behalf is still low, underscoring the adoption barrier that trust in AI represents.
For product marketing teams, this growing trust gap is both a challenge and an opportunity. We sit between Product, Marketing, Sales, Customer Success, Engineering, Design, and Customers, shaping the stories that determine how innovation is perceived.
In the AI era, our role now includes safeguarding the integrity of how (the use of) AI is communicated, implemented, and governed across the GTM journey.
PMMs must evolve into the ethical gatekeepers of AI‑powered GTM, ensuring that growth does not outpace integrity. And we are the architects of trust.
The trust crisis in today’s GTM
Artificial intelligence has become a powerful accelerant in modern GTM. From content generation to intent modeling, AI enables marketers to operate faster than ever. Yet as adoption accelerates, so does skepticism: speed and trust often move in opposite directions.
Harvard Business Review notes rising unease about opaque algorithms, biased outputs, and misuse of personal data. McKinsey’s guidance on explainability underscores that clarity about how AI influences decisions is not just a technical nice‑to‑have; it is essential communication that builds confidence for customers, partners, and regulators.
This distrust shows up across GTM execution:
- AI‑washing: Overstating a product’s AI capabilities to appear more advanced than it is. This erodes credibility when buyers discover the exaggeration. (Let’s be honest, we are all tired of hearing every brand claiming to be an AI product or AI-powered.)
- Hallucination risk: Generative AI models can fabricate facts or misstate product features, leading to misinformation in public assets if not reviewed.
- Opacity: When people cannot tell where AI is used or why they are seeing a certain message, authenticity about the brand and product declines, and skepticism grows.
When customers sense automation without accountability and authenticity, trust evaporates faster than any funnel metric can capture.
For product marketers, this is not a theoretical risk. It is a measurable business problem. Declining trust reduces conversion, word‑of‑mouth, and LTV, which no volume of AI‑driven efficiency can offset.
The brands that will win the next era of marketing will move from automation at all costs to authenticity at every touchpoint.
Product marketers as trust architect
If the past decade of modern product marketing was about mastering growth, the next decade will be about mastering trust.
As marketing teams embed generative and predictive AI into their GTM stack, the product marketing function is evolving from storyteller to trust architect: the person who ensures that what is built, promised, and delivered all align under a single truth.
We occupy a unique vantage point between product, brand, sales, and customer experience. This intersection gives us a rare ability to sense when a claim feels inflated or hyperbolic, when automation overshadows authenticity, or when an output crosses the ethical boundary between persuasion and manipulation.
In many ways, PMMs are the quality assurance layer for credibility in an age of automated communication, filled oftentimes with noise more than authentic voice.
McKinsey’s work on explainability highlights that this is not just a data science concern; it is a communication task that helps business leaders “confidently convey AI decisions to customers and regulators”.
In practice, being a trust architect involves three core responsibilities:
Verification
Before amplifying AI‑generated insights or copywriting, validate accuracy and source integrity. Harvard Business Review warns that unverified outputs can introduce misinformation at unprecedented speed.
(P.S. I have to admit, I personally feel both excited and concerned when watching how ultra-realistic videos generated from Sora 2 were. Just imagine the risks it could introduce when those videos are turned into video campaigns.)
Transparency
Clarify where and how AI contributes to customer‑facing marketing assets. Bain’s research suggests transparency increases comfort and willingness to adopt AI‑enabled experiences, which may ultimately result in conversion.
Alignment
Ensure every AI‑powered message reinforces brand positioning rather than diluting it. As product positioning expert April Dunford would say, positioning is not what you say – it is what people believe. And AI-generated content should strengthen, not blur, that belief.
TAG framework: Guardrails for responsible and ethical AI in GTM
Ethical AI in GTM is not a corporate slogan. It is a set of habits that let marketing and GTM teams move quickly without breaking trust.
Below is a framework that I named “TAG”, outlining 3 guardrails written for a Product Marketer’s day‑to‑day work.
1) Transparency (for explainability)
For PMMs, explainability is simple: if AI helped you decide what to say or who to say it to, briefly show what went into the AI model and why the output makes sense.
It is not a math proof. It is a simple, short note that connects inputs to a recommendation in language a customer or executive would understand. Stronger explainability increases confidence among buyers and regulators.
To make this practical:
When AI shaped public‑facing copy in a meaningful way, add a light disclosure such as “drafted with AI and reviewed by our team”, and keep a small source appendix for any stats that made it into the asset.
Slack Terms of Service and Privacy language clarification (2024)
Following user backlash about whether customer data was used to train Slack’s AI, Slack clarified its policy and updated its public language.
Coverage noted the company’s commitment that it does not train generative AI models on customer content without opt-in consent, alongside clearer privacy principles and trust documentation.
For Product Marketers, the lesson is simple: your trust page and terms of services are GTM assets. Plain-English disclosures and specific opt-in/opt-out paths reduce churn and PR fire drills.
2) Accountability: AI claims integrity
If you write “our AI does X,” it should do X for real customers, not only in a perfect demo scenario. Treat AI language like regulated claims: specific, supportable, and clear about limits.
The U.S. Federal Trade Commission has warned and taken action against deceptive AI claims, which makes it essential to implement a simple substantiation habit before launch. Here are simple steps you can today:
- Keep a one‑page claims log that pairs each AI‑related statement with evidence: test results, docs, customer proof, and any important caveats.
- Ban or at least regulate AI‑washing internally by defining what is truly AI in your product versus rules‑based logic or human workflow, then align Sales and Customer Success on the same definitions.
- Add a pre‑flight check: “Are we overstating? Does this sound hyperbole? Are limits visible where a buyer would reasonably expect them?”
- Name an owner who can update or pull copy within hours if new facts emerge.
FTC crackdown on deceptive AI claims (2024)
In September 2024, the U.S. Federal Trade Commission announced a coordinated crackdown on deceptive AI claims and schemes, including actions related to fake-review tools, “AI lawyer” services, and get-rich-quick promises.
The agency also warned that quietly revising terms to broaden data use can be unfair or deceptive. For PMMs, this underscores the need for a pre-launch substantiation habit and fast correction paths post-launch.
3) Governance: lifecycle controls
Governance is how good intent becomes repeatable practice. You do not need a heavy program to start. A short monthly cadence and a few shared artifacts create the audit trail that buyers, analysts, and policy teams increasingly look for.
International bodies now publish practical scaffolding that you can adopt in steps, such as the NIST AI Risk Management Framework and ISO/IEC 42001, the first AI management-system standard.
To operationalize:
- Hold a 30 to 45 minute monthly AI usage review with product, legal, security, and marketing teams, discuss changes to models or data, known risks, upcoming launches, and mitigation plan.
- Define a lightweight RACI chart for AI-assisted content, such as who approves claims, who monitors performance and bias signals, who owns corrections after publication
- Keep three living artifacts: a claims log, a brief note of inputs and rationale for AI-assisted decisions, and a short limits and assumptions note.
In the launch room, these guardrails work together. Transparency tells the story of how a decision was made. Accountability proves the claim belongs in market. Governance makes the process repeatable at scale. And you can try out the TAG Framework regardless where you are with the product development lifecycle.
Where we go from here
At ALL IN 2025, Evan Solomon’s point was simple and powerful: trust accelerates adoption. When people feel safe with AI, new products spread faster. The inverse is also true.
Where trust is weak, even the most sophisticated GTM engine stalls. That is why product marketers need to treat trust as a first-class input to marketing strategy rather than an afterthought in a FAQ or a opaque footer content.
The AI era changes the shape of GTM, but it does not change human expectations. Buyers still want to understand what they are seeing, why they are seeing it, and what happens next if they choose you. The practices in this essay are not abstract.
They are habits any PMM can apply: show your work in simple terms, substantiate every AI claim, and create a light governance rhythm that repeats under pressure. These habits do not slow growth.
They make growth durable by preventing rework or failed product launches, reducing legal fire drills, and earning the credibility and trust that compounds product growth efforts.
The next wave of product adoption will be won by teams that move quickly without asking customers to take a leap of faith whenever they see or sense the use of AI. That is the opportunity for product marketers right now.
Be the translator who makes AI-powered experience understandable, the editor who keeps the claims about the use of honest, and the architect who turns good intent into repeatable practice.
If trust is the fuel of adoption, product marketers are the ones who keep the tank full.
