Adobe Firefly Custom Models: A Game Changer for On-Brand Visual Content

Thursday, 19 March 20267 min read1 views

Adobe has launched a public beta for Firefly Custom Models, enabling users to train AI image generators with their proprietary assets. This innovation allows for the creation of AI-generated visuals that consistently adhere to specific brand styles and aesthetics, offering unprecedented control over creative output. (Source: The Verge, 19 March 2026)

What Happened

•Adobe introduced Firefly Custom Models in public beta, allowing users to train the AI with their own image datasets.
•This new capability enables the AI to learn and replicate specific artistic styles, aesthetics, and character designs.
•The goal is to ensure generated images maintain a consistent, on-brand look and feel.
•Users can feed their proprietary images into the system to customise the AI's output.
•This represents a significant step towards more controlled and brand-aligned AI content generation.

Why It Matters for NZ Marketers

•NZ brands can now maintain visual consistency across all AI-generated marketing materials, crucial for brand recognition in a competitive market.
•Small to medium-sized NZ businesses (SMBs) can leverage this to produce high-quality, branded visuals without extensive design resources.
•Agencies serving NZ clients can offer more bespoke and scalable creative solutions, reducing manual design time.
•It democratises sophisticated visual content creation, allowing local marketers to compete with larger international players.
•Cost efficiencies in content production could free up budget for other marketing initiatives in the NZ market.

Strategic Implications

•Investigate integrating custom AI models into existing content workflows to accelerate visual asset creation.
•Develop clear brand guidelines for AI training data to ensure outputs align perfectly with brand identity.
•Explore new content formats and campaigns that were previously cost-prohibitive due to design demands.
•Upskill marketing teams in prompt engineering and AI model management to maximise the utility of these tools.
•Evaluate the potential for creating hyper-personalised visual content at scale for diverse NZ audience segments.

Future Trend Signals

•The rise of hyper-personalised, AI-generated visual content tailored to individual consumer preferences.
•Increased demand for proprietary data sets and brand-specific AI models as a competitive advantage.
•A shift in creative roles towards AI oversight, curation, and strategic direction rather than manual creation.
•Further integration of AI across the entire marketing tech stack, from ideation to execution and analysis.

Sources

Adobe’s AI image generator can now be trained on your own art — The Verge

Share this analysis

Help NZ marketers stay informed

Editorial note: This analysis is original, AI-assisted editorial content. All source material is attributed with links. No full articles are reproduced. Short excerpts are used under fair dealing principles.