Zuckerberg’s $500 Million Biohub AI Push Comes as Meta Pays Record‑Low 3.5% Tax Rate

TL;DR: The Chan Zuckerberg Biohub has earmarked $500 million for AI models of human cells, a move that coincides with Meta’s 2025 federal tax rate of just over 3.5% of revenue, its lowest on record.

The $500 million injection will fund a five‑year program to build predictive AI models that simulate human cell behavior. $400 million is allocated to the Biohub’s own artificial‑intelligence development, with the remainder supporting external research teams. The goal is to create tools that help scientists map cellular interactions across whole organisms, potentially accelerating disease‑cure research.

Alex Rives, head of science at the Biohub, warned that current data volumes fall far short of what accurate biological AI requires. “To build artificial intelligence that can accurately represent the full complexity of biology and accelerate scientific research, we need orders of magnitude more data than exists today,” he said, emphasizing the need for new observation technologies from molecular to tissue scales.

In the same fiscal year, Meta reported a federal income tax payment equal to just over 3.5% of its total revenue, the lowest rate the company has ever recorded. With $79 billion in profit for 2025, the effective tax rate translates to roughly $2.8 billion, far below the 21% statutory corporate rate. The gap of about $13.7 billion highlights a stark contrast between the company’s tax contribution and the scale of its new philanthropic investment.

The juxtaposition raises questions about corporate fiscal strategy and philanthropic priorities. While the Biohub’s funding represents a fraction of Meta’s earnings, it signals a shift toward high‑tech health research that mirrors earlier billionaire initiatives in the sector. Critics note that the tax savings could have supported broader public health programs, yet the investment may also generate breakthroughs that benefit society in the long term.

What to watch next: the Biohub’s progress on data acquisition and model validation, and any policy discussions on corporate tax rates for high‑profit tech firms.