Maintained by: Luciana Coutinho — Master Formulator, Chemical Engineer, founder of LULUBLE and SknSprk, PhD candidate in Quantum Biology and Natural Medicine.

Companion document: AAGF v.2026 AI Interaction & Ethics Disclosure

Why this protocol exists

Most public AI ethics disclosures are legal compliance documents. They satisfy regulatory expectations without explaining how AI is actually integrated into the work.

This page is different. It describes the operating framework — what I call the LULUBLE Alchemy AI Protocol — that governs how artificial intelligence is integrated across LULUBLE, SknSprk, LULUBLEacademy, and lucianacoutinho.bio.

This is published openly because radical transparency about AI use is part of the work itself. The integration of laboratory rigor and bio-energetic awareness — LULUBLE Alchemy — extends to how I integrate human judgment with machine capability.

The principle

LULUBLE Alchemy holds that Logos (rigor, chemistry, structure, science) and Mythos (frequency, coherence, intuition, soul) are not separate inquiries. They are two languages describing the same architecture.

Applied to AI integration, this means:

  • AI tools are instruments, not authors.
  • I am the accountable human, the voice, the judgment, the embodiment.
  • The tools handle scale, synthesis, structure, and execution at the substrate level.
  • I handle direction, calibration, voice fidelity, scientific judgment, and intuitive read at the architecture level.

This is the same discipline I apply to formulation: the raw materials and the equipment do their work; the formulator directs the integration. AI is the new equipment in the laboratory of the work I do publicly.

What AI is used for in this work

I integrate AI assistance across four functional layers:

1. Research synthesis

AI tools help me synthesize peer-reviewed papers, regulatory documents, ingredient databases, scientific literature, and adjacent fields (quantum biology, biophysics, neuroscience) into integrated briefings. The depth of research synthesis available through AI is significant; I direct what is researched and review what is synthesized.

2. Content generation (educational, editorial, strategic)

AI tools help me draft, structure, and refine written content — articles, course material, educational scripts, sales copy, technical documentation. Every published piece is human-reviewed against editorial discipline. The voice signature, framework integrity, and final approval are mine.

3. Video production

AI-assisted voice (Judith) narrates select educational videos. AI editing tools assist with transcription, audio cleanup, and structural editing. Visual generation tools support the visual layer. All output is human-reviewed before publishing.

4. Operational support

AI tools assist with file management, structured data, audit checks, and administrative tasks. Customer-facing AI chat is limited; human review applies to all consequential interactions.

What AI is NOT used for in this work

Equally important to declare:

  • AI does not make scientific or formulation decisions. Product development for clients (LULUBLE Launchpad, PD Brief work) is conducted under my direct judgment as Master Formulator with 25+ years of experience and 1,000+ products developed at Avon, Reckitt Benckiser, and Prestige Consumer Healthcare.

  • AI does not provide medical, dermatological, or therapeutic advice. Educational content about quantum biology, bio-energetic coherence, or skincare is informational. It is not a substitute for qualified professional consultation.

  • AI does not execute payment, financial, or contractual transactions. All commerce is handled through PCI-compliant providers (Stripe, Square, Shopify Payments) with human authorization at every step.

  • AI does not make claims I have not verified. Specifically: any assertion about ingredient safety, regulatory compliance (MoCRA, FDA, FTC), product efficacy, or scientific mechanism is verified against current regulatory standards and peer-reviewed literature before publication.

  • AI does not replace embodied teaching. When I teach (LULUBLE Vibrational Authority Beta Cohort, masterclasses, workshops, live sessions), the teaching is embodied. The work is direct. AI may have supported preparation; it does not deliver the live transmission.

The accountable human

The accountable human for all content, research, recommendations, and decisions made across LULUBLE, SknSprk, LULUBLEacademy, and lucianacoutinho.bio is:

Luciana Coutinho Master Formulator (1,000+ products developed) Chemical Engineer Founder, LULUBLE and SknSprk Advisor, Potion AI (cosmetic deformulation) PhD candidate, Quantum Biology and Natural Medicine 25+ years across Big Beauty (Avon, Reckitt Benckiser, Prestige Consumer Healthcare)

Human review is applied to every published artifact. AI-assisted content is reviewed against:

  • Voice fidelity (consistency with the LULUBLE Alchemy framework)
  • Scientific accuracy (against current peer-reviewed literature)
  • Regulatory alignment (MoCRA, FDA, FTC, EU Cosmetic Regulation)
  • Editorial discipline (clarity, precision, appropriate register)
  • Bio-energetic coherence (alignment with the deeper architecture of the work)

Editorial discipline applied to AI-assisted content

Every published piece passes through specific gates regardless of whether AI assisted in production:

Gate 1 — Voice fidelity

Does this sound like LULUBLE Alchemy work, or like generic content? Does it preserve the integration of Logos and Mythos, or does it collapse to either pure science or pure spirituality alone?

Gate 2 — Banned phrase check

Specific phrases that violate the editorial framework are filtered out — including marketing-coded language, regulatory-risky claims, and phrasings that would undermine the precision of the work.

Gate 3 — Scientific accuracy

Claims grounded in chemistry, biology, regulatory science, or quantum biology research are verified against peer-reviewed sources or established frameworks (Popp's biophoton research, Pollack's coherent water domains, McFadden's quantum tunneling, etc.).

Gate 4 — Regulatory alignment

Cosmetic claims are evaluated against MoCRA structure-function requirements. Supplement-related claims (when discussed) are evaluated against DSHEA. Educational claims are clearly distinguished from therapeutic claims.

Gate 5 — Final embodiment

Before publication, I ask: would I say this aloud, in my voice, in front of an audience of peers? If not, it is rewritten until I would.

Why this matters for the work

The LULUBLE Alchemy framework holds that integration is the discipline. Not separation. Not pretending. Not hiding the substrate while presenting only the polished surface.

Disclosing how AI is integrated into the work is part of the integration. Audiences who care about substantive depth deserve to know how the work is produced. Founders evaluating my consulting offerings deserve to know that AI tools support — but do not replace — Master Formulator judgment. Course participants deserve to know that the live transmission is embodied even when supporting materials are AI-assisted.

This protocol is published openly so that:

  1. There is no ambiguity about my AI use across public surfaces.
  2. The accountable human (me) is named.
  3. The specific functions where AI assists (and where AI does not assist) are specified.
  4. The editorial discipline applied to AI-assisted content is documented.
  5. Visitors can evaluate whether this approach to integration aligns with their expectations.

Evolution

This protocol will be revised as AI tools evolve, as disclosure norms shift, and as my own integration of AI matures. The current version is v1, May 2026.

When the protocol is significantly revised, the version number will increment and the prior version will be archived for reference.

Companion documents

Closing

The work is patient. The discipline is the framework. The integration is non-negotiable.

AI is the new equipment in the laboratory. I remain the formulator.

ADSUM.

— Luciana Coutinho LULUBLE Alchemy AI Protocol v1, May 2026