Key Takeaways:
- AI design tools are now production-grade: Claude Design and Google’s Stitch can generate full pitch decks, prototypes, and landing pages from a brief. Your team is already using them, with or without sanction.
- md is the new format that ties them together: Open-sourced by Google Labs in April 2026, it gives AI tools a portable, machine-readable description of your brand. It is the first widely adopted standard for an AI-readable brand system.
- What it really signals is a higher bar for brand work: Brand systems now have to be documented well enough that a machine can apply them consistently. Loose Figma libraries and PDF guidelines no longer cut it.
- For biotech, the implications are sharp: A brand system has to read clearly to scientific audiences and translate cleanly into structured data. The judgment that goes into the system matters more than ever — because AI will faithfully reproduce whatever you give it, good or bad.
What is DESIGN.md?
A DESIGN.md is a single markdown file that describes a brand identity in two layers. The top of the file is YAML: machine-readable design tokens for colors, typography, spacing, and components. The bottom is markdown prose explaining the rationale: why these colors, why this type system, how to apply it.
The tokens give an AI agent exact values. The prose tells it why those values exist. The format includes validation tools — contrast ratio checks, broken reference detection — and exports cleanly to Tailwind and the W3C Design Tokens spec, which means that the same file can feed both your AI design tools and your production website.
That sounds technical, but the underlying idea is simple: a brand system, written down well enough that a machine can apply it without supervision.
Something Quietly Important Is Happening to Brand Design
Across most growth-stage biotechs, AI is already producing a meaningful share of the visual work. Founders draft Series B decks in Claude. Product managers spin up feature mockups in Stitch. Marketers generate landing pages and one-pagers in tools that did not exist eighteen months ago.
None of those tools share a common understanding of what your brand actually looks like. They infer, they guess, and they pull from defaults. The result is brand drift — across surfaces that you did not plan for, faster than your design team can catch.
In April 2026, Google Labs open-sourced DESIGN.md, a format spec built to close that gap. Around the same time, Anthropic launched Claude Design, which can generate complete creative work from a single prompt. We have spent the last few weeks reading both, thinking about what they mean for the biotech brands that we build, and watching how the conversation is shifting. This is what we have learned.
Why DESIGN.md Should Get a Biotech Marketer’s Attention
DESIGN.md itself is not the story. The story is what the format is forcing into the open: brand systems now have to be rigorous enough that a non-human can apply them consistently.
That is a quiet but real shift. For most of the last decade, biotech brand work has lived in a 60-page PDF guideline document, a Figma library that only the design team really understands, and a set of habits that get carried forward by whoever has been at the company longest. None of that is structured enough to feed an AI agent. As more of the day-to-day creative work shifts onto agents — in your marketing team, on the founders’ own laptops, and even inside the BD function — those loose systems start to fall apart.
The question moves from “Does our brand look good?” to “Is our brand documented well enough that a thousand AI-generated outputs all look the same?” That is a very different bar.
The Life Science Wrinkle
Biotech brands carry constraints that most consumer brands do not. Visuals have to read clearly to scientific buyers without dumbing down or losing technical accuracy. Type systems have to hold up in dense data visualization. Color has to survive grayscale poster printing at industry events like AACR. Brand voice has to signal credibility to technical audiences and investors at the same time. A booth panel reads differently than a website, which reads differently than a peer-reviewed figure. The list goes on and on.
None of that is a default that an AI tool can infer. None of it lives in a token file by accident. It comes from real design judgment, applied by a team that has watched what works and what does not in front of life science audiences.
Which gets to the part that does not get said enough: an AI design tool will faithfully reproduce whatever brand system it is given. If the system is sharp, every output is sharp. If the system is mediocre, the AI will scale the mediocrity across a thousand assets. The format is just a container. The judgment that fills it is what determines what comes out the other side.
What This Means in Practice
Here are a few honest takeaways for marketing leaders watching this from the inside:
- Audit how AI is already being used in your org. Your team is generating decks and collateral with AI tools today. The first step is knowing where, by whom, and to what end.
- Stop treating brand guidelines as a static PDF. Whatever format you land on — DESIGN.md, a structured Figma library, or something else — your brand system needs to be documented at a level of rigor that holds up under repeated automated use.
- Spend the design budget on judgment, not just deliverables. The teams that win the next two years of biotech brand work will not be the ones with the prettiest deck. They will be the ones whose brand systems hold up across a hundred AI-generated outputs without anyone having to step in and clean up afterward.
Where Samba Scientific Sits in This
We are not selling a DESIGN.md service (yet). The format is too new, the tools around it are still maturing, and we would rather show our work than oversell it. What we will say is this: the kind of brand work we have been doing for life science companies — conference systems, websites, presentation templates, rebrandings, scientific collateral — is exactly the kind of work this new era rewards. Rigorous brand systems, built with biotech audiences in mind, designed to hold up across whatever surface they end up on next.
If your brand is heading into a refresh, a relaunch, or just needs a serious second look, we would be a good team to talk to. You can see the work on our design services page, browse what we have built across the industry in our case studies, or book a 20-minute call to walk through what your brand looks like today and where it might need to be a year from now.
