The AI that learns you.
Adobe already has the data. No competitor does. The Firefly Personal Style Engine turns years of individual creative behaviour — sitting unused across every CC app — into a personalisation engine that makes every creator's AI feel like it was built specifically for them.
Right now Firefly generates from generic prompts. It doesn't know who you are, how you work, or what your creative style looks like. Every creator starts from zero every single time.
The Firefly Personal Style Engine fixes that. It's an AI layer that sits inside Creative Cloud and passively watches how you work — your colour choices in Lightroom, your compositions in Photoshop, your design decisions in Illustrator.
It builds a personal creative fingerprint from that behaviour automatically, without any extra work from you. Over time, every Firefly generation starts reflecting your specific aesthetic rather than a generic output.
The AI learns you. The longer you stay in Adobe, the more it sounds, looks, and feels like you made it.
The Style Engine doesn't launch everywhere at once. It starts where the creative signal is strongest — in the apps that dominate professional design workflows and capture the highest density of style-defining decisions.
Photoshop, Illustrator, and InDesign represent the core of industry adoption across graphic design software, based on company usage data and market share analysis (2025–2026). Together, they cover image editing, vector design, and layout — the three pillars of modern creative production.
Every edit, every colour choice, and every compositional decision made inside these apps becomes a data point that the Style Engine learns from — building a personal creative fingerprint that becomes more refined with every session.
The primary source of style signal for the fingerprint. Every tone curve adjustment, every colour grade, every compositing decision is a data point. Photoshop is where most creators express their visual identity most completely — making it the richest training ground for personalisation.
The fingerprint layer for graphic designers and illustrators. Illustrator captures compositional instincts, typographic preferences, and colour palette decisions that define a designer's personal visual language. The signals here are more structural — how a creator builds and organises space on a canvas.
Currently the most underserved app for AI in the Adobe suite — which makes it the biggest opportunity. Magazine designers, publishers, and layout professionals have a deeply personal approach to grid systems, hierarchy, and typographic rhythm. The Style Engine brings personalised AI to a workflow that has had almost none.
The Style Engine starts in these three apps because the data is already there — years of individual usage sitting uncollected. Phase 5 extends the fingerprint into Premiere Pro and After Effects, meaning every creative discipline inside Adobe eventually benefits from the same personalisation layer.
Every competitor is working from a snapshot. Adobe's fingerprint works from years of continuous creative behaviour.
Midjourney builds style personalisation through reference images you manually feed it. It sees outputs you show it — not the thousands of micro-decisions you make while actually editing.
Canva's AI personalisation is brand-kit level — your logo, your colours, your fonts. That is template consistency, not personal style intelligence.
These tools offer AI-powered style presets — static filters you pick and apply. No learning over time, no observation of habits, no compounding intelligence.
Custom Models let you upload images to train Firefly on a specific style. Manual, deliberate, static — you do the work, you get a model for that one specific thing.
Sensei is task intelligence — it helps you complete a specific action faster. Every Sensei feature is reactive and triggered by a task in the moment.
Every competitor is working from a snapshot — a reference image, a brand kit, a preset you chose. Adobe's fingerprint works from years of continuous creative behaviour across Photoshop, Illustrator, and InDesign.
Adobe has powerful tools. None of them do what the Style Engine does.
Helps you complete specific actions faster — subject selection, background removal, reframing. Every feature is reactive and triggered by a task you're performing.
Upload a set of images to train Firefly on a specific subject or style. One-time, deliberate, and static — you feed it references, you get a fixed model.
Syncs your company's colours, fonts, and logos across team members. Designed for marketing teams maintaining consistent brand guidelines.
The data exists across Photoshop, Illustrator, and InDesign. Sensei provides the ML infrastructure. Firefly provides the generation pipeline. The Style Engine connects two systems Adobe already owns that don't currently talk to each other.
Midjourney, Canva, and Luminar all work from snapshots — a reference image, a brand kit, a preset. Adobe's fingerprint builds from years of continuous behaviour across multiple professional apps. That foundation cannot be copied.
$390,000 to build against up to $396 million in annual revenue upside. Year 1 cost represents 0.01% of Adobe's AI budget. Even the conservative 2% upgrade scenario returns $79M annually — a 200x return in year one.
The longer you stay in Adobe, the more the AI sounds like you. Switching platforms means losing your creative memory. That retention effect compounds every year a creator stays — making churn increasingly unlikely over time.
For the first time, a creator's AI doesn't just generate — it reflects their personal aesthetic. The Style Dashboard gives full visibility and control over your creative fingerprint. Transparency builds trust. Trust builds loyalty.
Three phases over 12 months mean Adobe can validate at each stage before committing to the next. The first 2 months require only instrumentation — no generative changes, no product launches. Lowest possible risk entry.