The HEARTS Framework
The HEARTS Framework is a six-part method for evaluating whether an AI-assisted research workflow is actually better — not just faster — before you trust it. Developed by me (Kaleb Loosbrock) with the AIxUXR community, it provides UX and product research teams with a shared language for determining whether an AI-assisted process is rigorous, ethical, and worth its name.
HEARTS is how we, as UX Researchers, incorporate AI into our workflows with intention and integrity.
What does HEARTS stand for?
HEARTS is an acronym for six design principles you run an AI-assisted research workflow through:
H — Human-Led & Centered
E — Experience-Focused
A — Amplification, Not Automation
R — Rigorous & Responsible
T — Trustworthy & Transparent
S — Safe, Secure & Sustainable
Each letter is a lens, and each lens comes with one question you ask your workflow.
H — Human-Led & Centered
The researcher is the pilot, not the passenger. AI is a tool directed by and for humans. We prioritize the needs, context, and well-being of all people in the research process — ourselves as practitioners, our participants, and our partners.
Ask: Where does the human stay in control? If the answer is "nowhere," you've built an automation, not an AI-assisted workflow.
E — Experience-Focused
Deliberately design for the experience of all. We are accountable for the quality of the experience. Every interaction with AI and its outputs — whether by a researcher, a participant, or a stakeholder — must be intuitive, respectful, and positive.
Ask: Does this improve the experience for everyone it touches, or just speed things up for you?
A — Amplification, Not Automation
AI amplifies researchers' superpowers — you just have to earn it. We use AI to augment our skills and amplify our impact, not to automate our craft and dilute our value. We delegate the repetitive, annoying, and unnecessary tasks, freeing our time and brainpower for what humans do best: critical and divergent thinking, strategic synthesis, and building empathy.
Ask: Is AI freeing the researcher to do something more valuable, or just doing the valuable part for them?
R — Rigorous & Responsible
Maintain integrity, minimize bias, maximize confidence. We are fiercely protective of our work, our conduct, and our outputs. We proactively mitigate bias and maximize accuracy and confidence in every tool we use and every process we design. We don't "check the box" on ethics; we build our practice around it.
Ask: How are you validating accuracy and checking for bias — your own and the model's?
T — Trustworthy & Transparent
Transparency is the bedrock of trust. Traceability and transparency are the bridge to trust. From data collection to analysis, we provide a clear path for verification and prove the lineage of our insights. We disclose by default to build participant and partner confidence.
Ask: If someone audited this process tomorrow, could they follow the trail?
S — Safe, Secure & Sustainable
Protect the people, the data, the practice, and the planet. Three layers:
Safe — the well-being of practitioners, participants, and partners.
Secure — a zero-trust standard for their data, so no personally identifiable information ever lands in a public model.
Sustainable — workflows your team can actually maintain, with an honest eye on environmental cost.
Ask: Are you protecting people's safety, securing their data, and building something that lasts?
How do you use the HEARTS Framework?
Before launching any AI-integrated study, score your workflow on each pillar from 1 to 5, where 1 is high-risk and 5 is high-integrity. Write down the gaps in plain language.
Then apply one rule: a score of 1 or 2 on Rigorous, Trustworthy, or Safe, Secure & Sustainable means you redesign before you collect data — not after. Those three are where the damage is hardest to undo. Run the scorecard again after the project ships to see whether the experience and the rigor held up.
How does HEARTS map to other Responsible AI frameworks?
HEARTS translates the established Responsible AI consensus and re-centers it for User Researcher workflows. It aligns with the major frameworks on most dimensions and adds two that the compliance frameworks miss — Experience and Amplification.
| HEARTS | OECD AI Principles (2024) | NIST AI RMF | EU AI Act | Academic consensus |
|---|---|---|---|---|
| Human-Led | Human-centred values | Human role in accountability | Human oversight (Art. 14) | Human oversight & control |
| Experience-Focused | Well-being (partial) | — | — | — (UX-specific) |
| Amplification | Inclusive growth & well-being (partial) | — | — | Human–AI complementarity (partial) |
| Rigorous & Responsible | Robustness; fairness | Valid & Reliable; Fair (bias managed) | Accuracy, robustness, risk mgmt | Fairness & non-discrimination |
| Trustworthy & Transparent | Transparency & explainability; accountability | Accountable & Transparent; Explainable | Transparency (Art. 50) | Transparency; accountability |
| Safe, Secure & Sustainable | Security & safety; sustainable development | Safe; Secure & Resilient; Privacy-Enhanced | Safety; data governance | Privacy & security; societal/environmental well-being |
Where did the HEARTS Framework come from?
HEARTS wasn't invented from a blank page—it was built from experience and secondary research on Responsible AI frameworks.
While leading foundational Responsible AI research at Instacart (practitioner interviews plus a review of the major reports and standards), I found that the OECD, NIST, the EU AI Act, and the academic literature circle the same core: keep humans in control, be fair, be transparent, be accountable, protect data, and don't harm people or the planet. However, that consensus was written for governments and engineering orgs—not for the researcher staring at a transcript wondering if the AI summary is trustworthy.
HEARTS translates and applies Responsible AI frameworks for the researcher doing the work, and adds the two things the compliance frameworks miss: the experience of everyone the research touches, and the difference between AI that amplifies you and AI that quietly replaces you.
The HEARTS framework was first publicly announced at the AI Club: The UXR AI Playbook webinar held on September 26, 2025. Since then, the framework has been refined and updated based on pressure-testing in enterprise workshops, the AI in UXR 101 course, and feedback from researchers in the AIxUXR community.
Frequently asked questions
What is the HEARTS Framework? HEARTS is a six-part framework for evaluating whether an AI-assisted research workflow is rigorous, ethical, and trustworthy — not just faster. It stands for Human-Led, Experience-Focused, Amplification, Rigorous, Trustworthy, and Safe, Secure & Sustainable.
Who created the HEARTS Framework? The HEARTS Framework was developed by Kaleb Loosbrock, a Staff/Principal-level UX researcher and AIxUXR consultant, based on his research and experience integrating AI into his workflows at Instacart, and was refined with feedback from fellow researchers in the AIxUXR community.
What is HEARTS used for? It is used to evaluate, audit, and design AI-assisted UX and product research workflows — scoring a process against six principles before a team trusts or ships it.
How is HEARTS different from the NIST AI RMF or the EU AI Act? HEARTS aligns with those frameworks but is built for the working researcher rather than for governments or engineering organizations, and it adds two dimensions they don't name: Experience and Amplification.
Does HEARTS replace Responsible AI standards? No. HEARTS translates and operationalizes the Responsible AI consensus (OECD, NIST, EU AI Act) for research practice; it complements those standards.
Does HEARTS cover accountability? Yes — by design, not as a separate letter. Accountability lives inside R (we own the rigor of our outputs) and T (we make our decisions traceable and disclosed). It is covered across the framework, not bolted on.
Cite this framework
The HEARTS Framework, developed by Kaleb Loosbrock with the AIxUXR community. https://www.heykaleb.com/hearts-framework