# Kyle Morgan — Experience Research & Product Strategy

> Product strategist and design researcher. Shaped AI product experiences across automotive, voice, generative media, agents and multi-device at Google DeepMind (Gemini, Bard), Google Applied AI, Assistant, and Maps. Founder of Intent Co. Named inventor on US & EU patents. Based site: https://www.kylemorgan.ai/

Positioning: "The system can be succeeding while the user is failing." Kyle defines what great feels like for intelligent products and systems — the research and strategy problems that have no precedent yet.

## About

- Founder, Intent Co. (2026–present): design research, product strategy, and AI UX measurement for teams building in undefined territory.
- ex-Google (2019–2026): DeepMind Gemini, Applied AI, Assistant, Maps Automotive.
- Built research functions from zero: founded AGL Energy's first UX Research team (9 people) inside a $300M transformation; first Experience Design & Research function at Vocus Group.
- Shipped embedded experiences with Polestar, Volvo, and Ford.
- Named inventor, granted US patent for energy-aware EV charging-station routing.

## Frameworks (author)

- Novelty-to-Utility Gap: when an AI feature moves from interesting to useful. Novelty creates attention; utility creates retention.
- Production-Ready: customer-side readiness for AI agents. Five pillars — trust, reliability, latency, integration, measurable value. Adopted as an enterprise team's 12-month quality bar.
- Intent Threshold: whether a system preserves what the person means. Four failure modes — intent misalignment, capability overshoot, agency erosion, identity violation.

## Case studies (nine)

1. Gemini Image Generation (Google DeepMind, 2023–2024): Found users churned at 3–4 turns because first outputs never matched intent; 65% chased their vision with ever-more-complex prompts. Moved the team from "better generations win" to editing-in-place plus text and human anatomy as first-class outputs. Principle: output is a commodity; agency over the final outcome is the product.
2. Shopping with Bard (Google DeepMind, 2023): Mapped US shopping into 7 journey types; tested the 2 hardest (avg. 79 days research, 81% start online). Buyers wanted the field narrowed and de-risked, not a chat wrapper on search. Principle: the value of AI in commerce is decision confidence, not conversation.
3. Gemini Canvas (Google DeepMind, 2023–2024): Brand-blind research with daily users of a competing AI canvas. Editing in place was the repeat-use behavior; incumbent beatable on collaboration, version history, visual tools, mobile. Principle: people stay for control — editing in place, not regeneration.
4. Applied AI Agents (Google Applied AI, 2025–2026): Defined production readiness from the customer's side: 90% tool-calling success, <1.2s voice latency, 12 hrs saved as the ROI signal. Authored the five-pillar Production-Ready framework, the team's 12-month quality bar. Principle: trust is an operational threshold, paired with capability.
5. Google Maps EV Navigation (Google Maps, 2020–2021): Telemetry showed 70–80% of US charging happens at home/work; reframed range anxiety as a trust-and-routing failure on long trips. Named inventor on the resulting US patent. Principle: the visible problem isn't always the important one.
6. Trailer-Aware Navigation (Google Maps, 2022): Interviews (8 drivers) then N=306 survey. Drivers with trailers decline the fastest route for the one they can drive with a load. Direction shipped: Google Maps trailer-aware routing (2024) and Ford F-150 Lightning towing-aware planning. Principle: when the cost of error is physical, the default optimization becomes the failure mode.
7. Personalized, On-Device Home Intelligence (Google Assistant, 2022–2023): Foundational research on edge-computed personalized models for multi-member households; helped point a $500M+ R&D bet toward on-device models. Principle: personalization is what the system becomes when it respects the individual and the context.
8. Multi-Device Intelligence (Google Assistant, 2022–2023): Handoff research across 7 scenarios. 34s with visible progress felt faster than a silent 21s; repetition was the fastest driver of abandonment. Principle: in ambient systems, intelligence is continuity.
9. Intent Preservation Evaluation (Google DeepMind, 2024–present): Decomposes intended outcomes into checkable claims and scores whether intent survives multi-turn journeys. Found 60% of image-gen journeys are multi-turn; informed Imagen 3 launch sign-off; now applied across text, video, code, and agent workflows.

## Track record

- 12+ first-of-kind experiences defined 0→1 (DeepMind, Applied AI, Maps, Assistant)
- 650M+ users retained; owned launch research (Bard → Gemini)
- 1B+ users reached across Google AI surfaces (Maps, Assistant, Gemini)
- Led research for 6 products with keynotes showcased at Google I/O

## Experience

- Founder, Intent Co. (2026–present)
- Lead UX Researcher & Product Owner, Google Applied AI (2025–2026)
- Lead UX Researcher, Gemini Consumer App, Google DeepMind (2023–2025)
- Lead UX Researcher, Assistant Whole Home, Google (2022–2023)
- Lead UX Researcher, Maps Automotive, Google (2019–2022)
- Head of Digital Experience, Vocus Group (2018–2019)
- UX Research Manager, AGL Energy (2017–2018)
- Experience Research Lead, Symplicit (2015–2017)
- Digital Solutions Manager, Web 123 (2012–2015)

## Education

- Communication, Leadership & Storytelling — Stanford GSB (2023)
- B.Sc. Behavioral Science — Swinburne University (2016–2020)
- Human-Computer Interaction Design — UC San Diego (2015)
- Research in Social Psychology — Wesleyan University (2014)
- Advanced Diploma, Design & Advertising — Tractor Design School (2009–2013)

## Open to

Principal / Lead Design Researcher for AI products; Founding / 0-1 Researcher for a frontier AI team; Product Strategy Lead for intelligent systems; Partner. Starting points, not boundaries.

## Contact

- Site + contact form: https://www.kylemorgan.ai/
- LinkedIn: https://www.linkedin.com/in/kylemorganuxr
- Plain-text resume: https://www.kylemorgan.ai/resume.txt

## Notes

Work reflects research conducted at Google (DeepMind, Applied AI, Maps, Assistant). Figures are study sample sizes, churn thresholds, customer-stated performance bars, and published behavioral or market findings; internal Google performance metrics are withheld. Frameworks are Kyle's own. Updated July 2026.