Arc Institute Announced as New Grantee in The Audacious Project's 2025 Cohort

Investment will support Arc's Virtual Cell Initiative, to build an accurate model of cellular responses to advance complex disease research

Arc Institute and The Audacious Project

Today, Arc Institute was announced as one of the newest grantees selected by The Audacious Project, the collaborative funding platform within TED. This selection recognizes Arc's bold effort to build and validate AI models that predict how cells respond to genetic, chemical, environmental, and metabolic perturbations across cell types and disease states.

The Audacious donor community has committed a total of $1.03 billion to this year's grantees, the largest raise to date and a significant infusion of capital into visionary ideas to shift systems and sectors. Arc is proud to be part of the 2025 cohort, which will advance medical breakthroughs and reproductive care, improve climate health, spur economic growth and strengthen education systems around the world.

"The Audacious Project grant allows us to really uplevel our Virtual Cell Initiative," said Silvana Konermann, Co-Founder and Executive Director of Arc Institute. "This is a multi-year project, beyond the scale that a single donor could support, and with potentially very large impact if it succeeds, which is exactly what Audacious was looking for. The funding is going to be genuinely transformative."

With this investment, Arc will pursue four interconnected goals:

Generate a scaled, diverse and high quality dataset. The biggest barrier to an accurate virtual cell is data. Arc's Technology Centers will conduct over one billion cell experiments, capturing how diverse human cell types respond to perturbations. Arc's dataset will use CRISPR to establish cause-and-effect relationships, span diverse cell types, and be scaled to the level required for AI models to meaningfully improve in their accuracy. We are also partnering with organizations like Tahoe Therapeutics and Biohub as well as 10x Genomics and Ultima Genomics, to accelerate our access to and ability to generate this training data.

Train and improve the model. Arc will use this dataset to train successive versions of its first virtual cell model STATE, releasing new versions regularly, with the goal to eventually create in silico results that closely match much slower-to-achieve experimental results. We will test multiple model architectures in parallel, such as the recently released STACK, to find the most effective path forward.

Validate in the real world. As a full-stack AIxBio research organization, Arc's labs will test virtual cell predictions against real experimental results. We will use Alzheimer's disease – an area in which our researchers have developed deep expertise – as a major internal test case. Alzheimer's is a classic complex disease, with many genetic and environmental risk factors posing a major unmet medical need. We will test our virtual cell model's ability to identify which interventions could reverse key Alzheimer's phenotypes, and validate these recommendations to better inform the models, creating a lab-in-the-loop to point toward potential drug targets.

Make it available. The full model will be published on GitHub and hosted through a free online portal for nonprofit researchers worldwide. Commercial licenses will be available to biotech and pharmaceutical companies.

Our Virtual Cell Initiative exemplifies Arc's unique architecture and willingness to invest in decade-scale fundamental mechanism work (like academia) combined with an eye toward translation (like industry), but without the timeline constraints of either. Our Technology Centers and Core Investigator labs are co-located, enabling research synergies and a tight feedback loop between experiment and computation. We leverage experimental techniques pioneered by our team to make causal, high-quality data generation possible at scale.

"This year's cohort represents what happens when humanity's toughest problems are met with courage. It is also a declaration of what they imagine is still entirely possible. Whether they are charting paths to plastic-free oceans, unlocking access to life-saving health services for women and girls globally, or harnessing AI to prevent disease, the breadth and depth of these interventions is awe-inducing," said Anna Verghese, Executive Director of The Audacious Project. "We're humbled yet again by what the Audacious community makes possible."

With support from The Audacious Project, past cohort members have redefined what is achievable in our world. Arc is honored and excited to be part of this legacy.

###

About Arc Institute

Arc Institute is an independent nonprofit research organization based in Palo Alto, California, that aims to accelerate scientific progress and understand the root causes of complex diseases. Arc's investigators are supported by long-term funding and freedom to pursue bold ideas. Its Technology Centers leverage multi-omics, genome engineering, and cellular, mammalian and computational models to advance discoveries at the intersection of biology and artificial intelligence. Founded in 2021, Arc partners with Stanford, UC Berkeley, and UCSF.

More AIxBio from Arc Institute

The Virtual Cell Initiative is one piece of a broader effort at Arc to build the full stack of interconnected AI and biology. On the AI side, these resources range from foundation models that learn the language of DNA, to virtual cell models that predict how cells respond to perturbations, to agentic mining of public data, to tools for training models and applying their predictions and designs. Learn more about:

  • Evo 2 — Arc's genomic foundation model and the largest AI model in biology to date, trained on 9.3 trillion nucleotides from over 128,000 genomes across all domains of life to read, write, and interpret the language of DNA.
  • State — Arc's first virtual cell model, trained on large perturbational datasets to predict how genetic, chemical, and environmental changes shift gene expression across cell types.
  • Stack — A single-cell foundation model that uses in-context learning to predict cellular responses to perturbations never directly measured. (Preprint)
  • scBaseCount — AI agents that find, clean, and uniformly process single-cell data for model training, part of Arc's Virtual Cell Atlas.
  • The Virtual Cell Challenge — An annual competition launched in 2025 to evaluate and improve virtual cell models across the field.
  • Multi-Evolve — An AI-guided protein engineering framework from the Hsu and Konermann labs that accelerates iterative design-test-build cycles.
  • CodonFM — A family of open-source AI models developed with NVIDIA that reveal the grammar underlying codon choice.
  • ProPer-seq — A cost-efficient method for linking perturbations to transcriptional phenotypes.

Subscribe to the Arc newsletter to stay up to date on our AIxBio work.

About The Audacious Project

Launched in 2018, The Audacious Project is a collaborative funding initiative dedicated to supporting bold ideas with the potential to change the world. Housed at TED and powered by the belief that ideas change everything, Audacious leverages a global network of visionary social entrepreneurs and funders to support bold solutions to the world's most urgent challenges. Audacious is issue-agnostic and supports grantees to balance audacity and achievability for lasting, systemic impact. Since 2018, Audacious has catalyzed more than $7 billion for 70 projects, demonstrating what's possible when we dream bigger and act together. The funding community includes respected donors and philanthropists including ELMA Philanthropies, Emerson Collective, MacKenzie Scott, Pivotal Ventures, Reed Hastings and Patty Quillin, Skoll Foundation, Valhalla Foundation and more.