AI Transforms Product & UX: New Frameworks for Excellence
AI accelerates development but risks mediocrity. A new framework, Sense, Shape, Steer, helps product and UX teams navigate these challenges and prioritize quality.
Key Insights
-
Insight
The boundaries between product, UX, and research roles are blurring, shifting the focus from 'who does what' to collaboratively determining 'what needs to be done and why' to solve problems effectively.
Impact
This fosters more agile, collaborative product development teams, leading to more user-centric solutions and efficient resource allocation in AI-driven projects.
-
Insight
AI's tendency to produce 'average mean' solutions creates a trap where readily available, high-fidelity prototypes can diminish human creativity and lead to widespread product mediocrity.
Impact
Highlights the risk of undifferentiated products if AI is used uncritically, emphasizing the need for human creativity and high standards to achieve market differentiation and user delight.
-
Insight
The accelerated speed of AI development is reducing industry patience for traditional discovery and research phases, making it challenging for researchers and designers to keep pace.
Impact
Underscores a critical challenge in maintaining user-centricity, potentially leading to 'build-first' mentalities that result in products lacking true market fit or deep user understanding.
-
Insight
A structured framework like 'Sense, Shape, Steer' is crucial for systematically approaching AI product development, moving beyond ad-hoc POCs to ensure AI solutions address real problems.
Impact
Provides a methodical approach for organizations to navigate the complexities of AI product development, reducing missteps and improving the success rate of AI initiatives.
-
Insight
Defining robust AI evaluation criteria ('AI evals') with subject matter experts is critical post-launch for sophisticated AI products, as traditional QA alone is insufficient to ensure accuracy, trust, and real-world utility.
Impact
Emphasizes that comprehensive, context-specific AI evaluation is crucial for successful deployment, preventing AI experiments from failing to deliver real-world value or causing unintended negative consequences.
Key Quotes
"The boundaries are actually blurring... It's less about who does what and more about really teams coming together and figuring it out in the sense what needs to be done and with that what needs to be done often irrespective of the role, really, be it the researcher, be it the designer, be it the product manager, the conversation often needs to start from what are we solving and why?"
"The way AI creates stuff, it is designed to be right... it is designed to be that average mean, which means you will never look at it and think that, oh my god, this is so bad, or no, this doesn't work. You will mostly actually nod your head and say that, okay, yeah, that makes sense. That that works. And that's really the trap."
"Steer is where you really define your AI evals, is one very important part of steer. And I have seen actually product making it or breaking it in terms of make or break kind of a situation for the products in terms of if they have got their AI events right or not."
Summary
AI Transforms Product & UX: New Frameworks for Excellence
The rapid advancement of AI is fundamentally reshaping the landscape of product management and UX design. While AI tools accelerate development to unprecedented speeds, they also present critical challenges: a blurring of traditional roles, a diminishing patience for deep discovery, and the pervasive risk of AI's "average" output leading to widespread mediocrity. For leaders in technology and business, understanding and proactively addressing these shifts is paramount to fostering innovation and delivering truly impactful solutions.
The Blurring Lines: Focus on 'What Needs to Be Done'
The era of rigid role definitions (researcher, designer, product manager) is giving way to a more fluid, collaborative environment. AI's ability to quickly generate prototypes and content means the focus must shift from "who does what" to "what problem are we solving and why." This necessitates interdisciplinary teams working in concert from the outset, ensuring that core user needs and business objectives remain the bedrock of development, even as building becomes faster. The danger lies in mistaking volume for quality; AI can produce seven pages of user stories, but the crucial question remains: "Is it good?"
The Trap of the 'Average Mean': Raising the Bar
AI's inherent design often steers solutions towards the "average mean." This creates a trap: readily available, high-fidelity prototypes can lead to diminished creativity, as teams may accept "good enough" rather than striving for exceptional, delightful experiences. In a world where AI can effortlessly churn out functional but uninspired designs, the onus is on designers, product managers, and researchers to actively "raise the bar." This means pushing beyond the default, challenging assumptions, and meticulously crafting solutions that truly differentiate and delight users. Without this critical human intervention, products risk becoming indistinguishable and failing to achieve meaningful adoption.
The Sense-Shape-Steer Framework: A Blueprint for AI Product Development
To combat these challenges and systematically approach AI product development, the "Sense, Shape, Steer" framework offers a structured methodology:
* Sense: This initial phase is about understanding what's possible and what's worth solving. It involves identifying user needs and AI capabilities, mapping out opportunities, and crucially, assessing risks and defining guardrails. Teams must collaboratively determine acceptable accuracy levels for AI (e.g., 90% for recommendations, 99% for critical functions like prescriptions) and consider the necessity of human-in-the-loop interventions. * Shape: Moving beyond interfaces, this stage emphasizes user-centric storyboarding. Teams should visualize the user's journey frame by frame, focusing on goals, actions, context, and precisely where AI can genuinely add value—whether by nudging, generating insights, or silently assisting—rather than simply automating. * Steer: The final, ongoing phase focuses on implementation and continuous evaluation. Defining robust AI evaluation criteria ("AI evals") is critical. This goes beyond traditional QA, often requiring subject matter experts to validate AI outputs against real-world accuracy and utility. Trust metrics and adoption rates must be carefully monitored, with preparedness for fallback options and human oversight, ensuring AI solutions deliver on their promise and don't just remain experiments.
Conclusion
The AI revolution demands more than just adopting new tools; it requires a fundamental shift in how teams collaborate, evaluate quality, and define success. By embracing frameworks like Sense, Shape, Steer, and prioritizing critical thinking over automated output, organizations can navigate the complexities of AI, build truly exceptional products, and secure a competitive edge in a rapidly evolving technological landscape.
Action Items
Implement cross-functional working sessions focused on 'what needs to be done and why' for new AI initiatives, consciously blurring traditional role boundaries between product, UX, and engineering.
Impact: Improves team alignment, fosters shared ownership, and ensures all efforts are directed towards solving validated problems, preventing misaligned priorities in AI development.
During the 'Sense' phase of AI product development, explicitly define desired AI accuracy levels, human-in-the-loop requirements, and fallback options based on the criticality and stake of the AI's function.
Impact: Mitigates operational and ethical risks, builds user trust, and provides clear, measurable targets for AI model performance, leading to more reliable and responsible AI solutions.
Prioritize user-centric storyboarding over immediate interface design when shaping AI experiences, focusing on mapping user goals, actions, context, and where AI genuinely adds value within the journey.
Impact: Encourages a deeper understanding of the user journey and AI's true value proposition, preventing the creation of unnecessary interfaces and ensuring AI integration is meaningful and impactful.
Establish comprehensive AI evaluation criteria ('AI evals') for deployed products, involving subject matter experts alongside QA, design, and engineering to validate accuracy, nuance, and real-world utility of AI outputs.
Impact: Ensures deployed AI solutions are thoroughly tested and perform as expected in complex scenarios, leading to higher quality products, increased user adoption, and reduced post-launch issues.
Actively challenge and push beyond 'average' or 'good enough' AI-generated output, empowering designers and product managers to 'raise the bar' for creativity, user delight, and unique problem-solving.
Impact: Fosters a culture of innovation and excellence, enabling products to stand out in a crowded market and deliver truly exceptional user experiences rather than bland, undifferentiated solutions.
Mentioned Companies
Coro UX Design
5.0The company's founder/CEO, Bancy Maida, developed a highly praised UX framework for AI that is central to the discussion.