The thinking in this film began in Down the AI Rabbit Hole. You can read it now, or save a quiet reminder here and return to it when you are ready.
Phillip Llewellyn built Human Heartbeat AI and the OSCAR Diagnostic after observing a consistent pattern across businesses adopting AI: the failure was not the technology. It was the absence of diagnosis before deployment.
His approach is governance-first, sequence-driven, and human-led. Human Heartbeat AI helps organisations and people work with AI without handing over the thinking that still needs to remain human. OSCAR, the Academy and bounded AI Worker roles each begin from a different practical question, but all return to the same point: accountable human judgement must remain visible.
Human Heartbeat AI is an emerging human-governed AI capability ecosystem—not a large consultancy and not a promise of automated outcomes. It is built around a single conviction: AI adoption done well requires clarity before tools, governance before automation, and human judgement before delegation.
Read the Founder’s Thesis →A tool is purchased before the operating position is understood. Automation is introduced before accountability structures exist. AI is deployed before anyone has asked what the business actually needs.
The result is AI adoption that creates noise, cost, and complexity rather than improvement. The tool is not the problem. The sequence is the problem.
Human Heartbeat AI was built to fix the sequence.
Every engagement begins with the OSCAR Diagnostic — a structured assessment that maps the operating position before any AI implementation is considered. Most providers start by asking what tool you want. We start by asking what is actually happening.
We are not a reseller. No vendor pays us a commission or referral fee, and no tool relationship shapes what we recommend. If the right answer is no AI at all, we say that. If the right answer includes a system we build ourselves, we say that too — but the recommendation always comes before the build, never the other way round.
AI Workers are introduced only after governance structures and Human Decision Gates are confirmed. Automation without accountability is not a feature. It is a liability.
Every AI system we design operates under a Human Decision Gate. A named human remains accountable for every consequential output. The system can be paused, overridden, or corrected at any point.
Every element of the OSCAR Diagnostic — the framework, the logic, the output structure, and the interpretation methodology — has been personally reviewed and approved by the founder before reaching a single client. What you receive has survived that scrutiny. The system does not run until it has.
Human Heartbeat AI works exclusively with UK businesses. The regulatory context, the operating pressures, and the commercial realities of UK SMEs and enterprise teams are the foundation of everything we do.
The book is where the thinking began. If you want to see how that thinking becomes a governed, practical approach to AI, explore why OSCAR comes first.
Why OSCAR comes first →