By Alice Gilman

In April, Health Secretary Robert F. Kennedy Jr. announced plans to dramatically reduce animal testing at U.S. health agencies. The FDA followed with a commitment to phase out animal studies in areas like monoclonal antibody development, favoring in vitro systems and artificial intelligence instead.

The move reflects growing momentum in both Washington and the biotech industry to leave behind animal models for good. It’s a goal I support not just as a scientist, but as a human being who believes we have a moral obligation to reduce suffering wherever possible.

But I also know this: we are not ready.

As Chief Operating Officer of a biotechnology company advancing ethically engineered biological models for drug testing and disease research, I operate at the intersection of research design, commercialization, and regulation. I’ve helped stand up lab infrastructure in frontier zones like Roatán, built operations for scientific events, and brought together the funding and logistics needed to pilot full-body NHP human-relevant alternatives.

I’ve worked in tissue engineering and regenerative medicine. I’ve helped build some of the most advanced organoid and organ-on-chip models currently used in preclinical research. I’ve seen first-hand what they can do and what they can’t.

Despite the hype, these systems still fail to replicate the physiological complexity of the human body. Organoids float in isolation, lacking the blood flow, immune signaling, or metabolic clearance that defines a real living system. Organ-on-chip devices can simulate shear stress or vascular flow, but only for isolated organ units. None of these platforms can yet recreate the interdependence that governs human biology.

It’s a limitation I’ve seen up close while helping investors evaluate next-gen biotech platforms. No matter how elegant the model, if it can’t capture whole-system dynamics, it doesn’t survive diligence.

And we’re paying the price for that fragmentation. According to a 2022 review in Nature Reviews Drug Discovery, fewer than 5% of drugs tested in organoid models progressed beyond Phase I clinical trials. That’s not a step forward.

The National Association for Biomedical Research responded to Kennedy’s announcement with an uncomfortable truth: “No AI model or simulation has yet demonstrated the ability to fully replicate all the unknowns about many full biological systems.”

They’re right. And those unknowns matter. We’ve seen promising drugs fail because a model didn’t account for liver metabolism. We’ve seen toxicities missed because we didn’t model immune response. If the goal is to protect human lives and accelerate development, we need better tools, ones that replicate not just parts, but the whole.

So what’s the path forward?

First, we need to stop pretending that replacing animal models with today’s in vitro tools is a one-for-one substitution. It’s not. These systems are incomplete and often lack external validation in humans. Relying on them too heavily, or prematurely risks pushing unsafe or ineffective drugs into clinical trials.

Second, we need to invest in the next generation of preclinical models: integrated, full-system human biology platforms. That means human cell-based models that incorporate vasculature, immune components, and endocrine signaling. Models that can metabolize drugs, develop inflammation, and respond systemically, not just in one tissue, but across many.

Yes, this will be difficult. Yes, it will take time. But it’s the only scientifically and ethically defensible path to eliminating animal testing. Anything less is wishful thinking.

And finally, we need to decouple the rhetoric from the reality. We cannot allow public policy to outpace the science. The FDA Modernization Act 2.0 opened the door to non-animal testing methods, and that’s a good thing. But regulation should not become a vehicle for political wins at the expense of scientific rigor.

I had once anticipated a life in consulting. I studied finance, management, international systems and expected to spend my career advising others through strategy decks and spreadsheets. But at some point, I realized that looking at numbers without acknowledging what they represent felt empty to me. A spreadsheet might show failure rates, clinical delays, or cost projections, but to me, those cells never felt neutral. Each one points to a decision that shaped a human life: a treatment not delivered, a risk not caught, a future lost too soon. I couldn’t build a career that required me to flatten those realities into metrics. I wanted to work somewhere that honored them and built systems that took them seriously.

This is personal for me.

I was born in Kazakhstan, where my grandmother regularly housed stray animals — sometimes over thirty at a time. Watching her care for each one with patience and dignity shaped me long before I understood what ethics or biology were. It taught me to see life as precious, even when it was inconvenient, unspectacular, or unseen.

As I grew older, I spent weekends helping local shelters and organizing small campaigns to sponsor vaccinations for street dogs and cats. But the root was always the same: a deep sense that every being is its own small universe and deserves thought, protection, and care.

That perspective stayed with me when I entered biotech. To me, reducing animal suffering isn't just a technical or regulatory challenge. It's a moral one. And building better models, ones that reflect the complexity and dignity of life is how we meet it.

Years ago, my father suffered a massive cardiac event. Four blocked chambers. Emergency surgery. Two weeks in a coma. That phone call changed the course of my life. I left policy consulting behind and turned to biology, believing we could build better models to understand and heal the human body.

But what I found were tools optimized for simplicity, not fidelity. Models built to mimic one organ at a time. Elegant in design. Devastating in their limitations.

The human body is not a collection of parts, it’s a system. We can’t keep studying diseases in pieces and hoping the results will scale. Whether we’re testing new drugs, mapping rare disorders, or training AI models, the biology we use needs to reflect the biology we live with.

I support the vision behind RFK Jr.’s announcement. I believe deeply that we must reduce and eventually eliminate animal use in research. But to do that responsibly, we have to acknowledge the gap between where we are and where we want to be.

That means saying out loud what many in science already know: we’re not ready. Not yet.

But we can be - if we build the right systems.

If we want to move beyond animal testing, we need to treat system-level modeling as national infrastructure. That means funding it like a public good, validating it like a regulatory standard, and building it with the urgency of a moonshot.

Alice Gilman is Chief Operating Officer of R3 Bio, a biotechnology company advancing non-animal models for ethical, system-level research. She previously worked in regenerative medicine, international operations, and commercialization of human-relevant biological platforms.