Insilico Medicine has introduced the first AI platform that models how cells age and respond to drugs over time, rather than treating them as frozen snapshots. The company's Virtual Aging Cell (VAC) platform, announced on August 14, 2026, represents a decade-long effort to bring biological time into computational drug discovery. Most existing AI models in drug development examine cells at single moments, missing the dynamic processes that actually determine whether a treatment will work in real patients.

Why Do Drug Companies Need to Model Aging Cells?

The fundamental problem with current virtual cell technology is that it treats biology like a photograph. Cells are captured at one point in time, but living cells are constantly dividing, differentiating, aging, and responding to their environment. This static approach misses critical information about how drugs interact with cells as they change over weeks, months, or years. By incorporating biological age as a core variable, Insilico's platform can simulate cellular differentiation, reprogramming, and aging, offering computational support for target identification, drug discovery, and geroscience research.

The VAC platform uses a multi-agent AI architecture, meaning multiple specialized AI systems work together to reason across six biological scales simultaneously. These scales span from the molecular level up through intracellular processes, intercellular communication, tissue environments, organ function, and organism-wide effects. This hierarchical approach mirrors how biology actually works, where changes at one level ripple through the entire system.

How Does Insilico's Approach Build on a Decade of Research?

The company's journey toward VAC began in 2014 when founder Alex Zhavoronkov posed an ambitious question at an NVIDIA conference: "Can NVIDIA help solve aging?" This sparked a research lineage that progressively expanded what AI could learn about biological systems. The progression moved through several key milestones:

• Early Target Discovery (2017): Insilico partnered with BioTime on a project called Embryonic.AI, which used deep neural networks to track cell differentiation during the embryonic-to-fetal transition. The team identified COX7A1 as a key transition factor, marking the first experimentally validated AI target discovery result and proving that AI could capture the molecular switches governing cell fate decisions.
• Precious1GPT (2023): This aging clock model integrated DNA methylation and gene expression data to predict biological age and distinguish disease from healthy samples. The model identified dual-purpose targets like APLNR and IL23R that showed both anti-aging and anti-disease potential.
• Precious2GPT (2024): This model advanced from prediction to conditional generation, creating synthetic multi-omics data with specific tissue and age characteristics. It outperformed baseline models like conditional GANs at predicting age in generated data.
• Precious3GPT (2024): A true multimodal transformer that integrated text, tabular data, and knowledge graphs across four species (mouse, rat, monkey, and human) and three omics modalities (transcriptomics, methylation, and proteomics). This model could perform cross-species age prediction, target discovery, and compound sensitivity prediction within a single framework.

"Every computational model needs a first principle, and ours is biological age. This has been a decade-long scientific journey, from raising an industry-defining question at NVIDIA GTC in 2014, to teaching AI the language of aging biology with the PreciousGPT series, and now reimagining the virtual cell through agentic AI swarms," said Alex Zhavoronkov, Founder, Co-CEO, and Chief Business Officer of Insilico Medicine.

Alex Zhavoronkov, Founder, Co-CEO, and Chief Business Officer of Insilico Medicine

What Makes This Different From Other Virtual Cell Approaches?

Virtual cells were named one of seven technology breakthroughs to watch in 2025 by Nature, and they have become a focal point for competitive research among top global teams. However, most existing models rely on data collected at isolated time points, capturing cells as static snapshots. Their capacity to simulate dynamic temporal processes and multi-scale interactions across biological hierarchies remains markedly constrained.

Insilico's VAC platform directly addresses this limitation by making biological age a core conditional variable rather than an afterthought. The multi-agent architecture allows different AI systems to specialize in different biological scales and then collaborate on reasoning about how changes propagate through the system. This approach acknowledges that a cell's state is shaped by intracellular molecular networks, intercellular communications, tissue microenvironments, and individual background parameters.

"Biology unfolds across time and interconnected levels, yet many computational models examine only one layer or moment in isolation. VAC is designed to bring those dimensions together, combining biological age with multi-agent reasoning," explained Petrina Kamya, Vice President, Global Head of AI Platforms and President of Insilico Medicine Canada.

Petrina Kamya, Vice President, Global Head of AI Platforms and President of Insilico Medicine Canada

What Are the Practical Implications for Drug Development?

The ability to model aging cells computationally could accelerate drug discovery for age-related diseases, which represent some of the largest unmet medical needs globally. Rather than relying solely on animal testing or early-stage human trials to understand how a drug affects cells over time, researchers can now run thousands of computational simulations to predict outcomes before entering expensive clinical phases. This could reduce development timelines and costs while improving the likelihood that a drug will work in real patients.

The platform is also designed to support geroscience research, which focuses on understanding the biological mechanisms of aging itself. By modeling how cells age under different conditions and interventions, researchers can identify new targets for drugs that might slow or reverse age-related decline. This opens possibilities for treatments that address the root causes of age-related diseases rather than just managing symptoms.

Insilico has launched a dedicated Virtual Aging Cell webpage and is previewing the multi-agent driven VAC generation platform for collaboration with global pharmaceutical companies and research institutions. The company is actively seeking partners to model virtual cells, virtual systems, and virtual populations across both healthy and diseased states.

" }