AI Ethics · Health Tech · Embedded Strategy
AI Readiness Diagnostic
Helping a health tech startup ask the hard ethical questions about their AI before scale made those questions a lot more expensive to answer.
The Context
An emerging health tech company had built real momentum around an AI-driven product, and they knew that was exactly the moment to bring in ethical, anthropological, and user-centered thinking, before the foundational design decisions hardened into place.
The Ask
I came in to assess how ethically, culturally, and experientially ready their AI product actually was, then turn that into strategic recommendations that kept it aligned with real user needs, trust-building, and responsible deployment in the real world.
That work directly changed how the product was positioned, shortened iteration cycles, and held back a few risky feature releases until they were actually ready.
The Work
I started outside the product, not inside it.
I explored the broader sociotechnical and cultural landscape around the product's domain, surfacing the ethical tensions, user concerns, and social expectations already in play before anyone built a feature.
Then I went looking for what people actually wanted.
I surfaced the motivations, fears, and desired outcomes of the target user groups, bridging what the developers intended with what users were actually experiencing.
I audited the product's assumptions.
I reviewed the AI model's projected use cases and user flows to find the gaps in empathy, usability, and inclusivity before they became real problems.
I advised on where the guardrails needed to go.
I delivered recommendations for embedding ethical safeguards directly into the product's design, communication, and rollout strategy.
And helped them change how they talked about it.
I helped shift the team's framing from "intelligent automation" to "relational augmentation," a shift that aligned the product with user trust and where it needed to go long-term.
The Outcome
Here's what changed:
Reduction in Iteration Cycles
Getting user needs right early meant the versioning process itself sped up, and this was the difference it made.
User Personas Delivered
Each one captured real attitudes and expectations toward the beta product, giving leadership something to ground strategy in besides assumptions.
AI Readiness Framework Established
This became the structure the team could return to whenever technical feasibility, user needs, and ethical considerations needed to be weighed against each other.