Biotech vs AI vs Climate Tech: How Research Momentum Differs Across the Three Largest Innovation Verticals

Investors often group AI, biotech, and climate tech under a single "deep tech" umbrella. That framing obscures more than it reveals. These three verticals differ in publication cadence, geographic distribution, citation dynamics, and the lag between research acceleration and commercial traction. The Finch Innovation Index tracks all three across dozens of sub-themes, and the momentum patterns diverge in ways that matter for capital allocation timing.

This post maps those structural differences and explains what they signal for investors operating on a 2 to 5 year horizon.

Publication Volume and Acceleration Rates

AI research produces the highest raw preprint volume of any vertical tracked by the Finch Innovation Index. AI-related preprints on arXiv alone exceed 150,000 per year, dwarfing biotech and climate tech output on comparable platforms. But volume alone is a poor signal. What matters is acceleration: the rate of change in output across specific sub-themes within each vertical.

AI sub-themes tend to exhibit rapid, spiky acceleration patterns. A new architecture or training method can drive a 3x to 5x increase in related preprints within a single quarter, then plateau as the community absorbs and iterates. Biotech research momentum is slower and more sustained. Sub-themes like mRNA therapeutics or CRISPR delivery mechanisms show steady annual growth rates in the 15 to 30 percent range, compounding over multiple years rather than spiking. Climate tech research acceleration sits between the two, with sub-themes like solid-state batteries and direct air capture showing consistent year-over-year growth but rarely the explosive surges seen in AI.

These different acceleration profiles have direct implications for how momentum scoring captures signal strength and how investors should interpret it. A momentum spike in AI may indicate a 12 to 18 month commercialization window. The same score in biotech likely points to a 3 to 5 year trajectory.

Geographic Concentration and Its Investment Implications

The three verticals also differ sharply in geographic research concentration. AI research output is heavily concentrated in the United States and China, with those two countries producing roughly 60 to 70 percent of top-tier AI preprints. This bipolar concentration creates a clear map for sourcing deal flow and anticipating regulatory dynamics. Climate tech research is more geographically distributed than AI, with significant output from the EU, South Korea, Japan, and Australia alongside the US and China. Biotech research concentration falls between AI and climate tech, with the US, China, the UK, and Germany as dominant producers but meaningful contributions from smaller nations with specialized strengths, such as Israel in immunotherapy and Singapore in synthetic biology.

For sovereign wealth funds and multinational corporate R&D teams, these geographic patterns shape where to position technology scouts and which bilateral research corridors to monitor. The Finch Innovation Index surfaces geographic patterns at the theme level, making it possible to track shifts in national research emphasis before they show up in patent filings or trade policy.

Citation Velocity and Translation Speed

Citation velocity, how quickly a preprint accumulates references from subsequent work, varies systematically across the three verticals. AI preprints exhibit the highest short-term citation velocity of the three verticals, often accumulating the bulk of their citations within 6 to 12 months. This reflects the field's rapid iteration cycle and lower barriers to replication. Biotech preprints accumulate citations more slowly but sustain citation relevance over longer periods, often 3 to 7 years. Climate tech citation patterns tend to mirror biotech more than AI, with slower initial uptake but durable long-term relevance.

This difference matters because citation velocity is a proxy for translation speed from research to application. AI's fast citation cycle correlates with its relatively short lab-to-product timeline. Biotech's slower cycle reflects regulatory overhead, clinical validation requirements, and longer development timelines. Climate tech faces its own translation bottlenecks: hardware demonstration, permitting, and infrastructure integration.

What This Means for Portfolio Timing

The structural differences across these three verticals argue against applying a single investment timing model to all deep tech. AI momentum signals reward faster response times; a six-month delay in acting on a research acceleration signal can mean missing the seed and Series A window entirely. Biotech momentum signals support longer deliberation; the window between research acceleration and investable company formation is wider, often two to four years. Climate tech sits in the middle, with hardware-intensive sub-themes requiring patient capital but software-adjacent areas like grid optimization moving on timelines closer to AI.

The Finch Innovation Index scores momentum across all three verticals using the same methodology, but interpreting those scores requires vertical-specific context. A momentum score of 8 in an AI sub-theme and a momentum score of 8 in a biotech sub-theme do not imply the same investment timeline. Understanding these translation dynamics is what separates systematic research intelligence from naive publication counting.

For a deeper look at how the index defines and scores themes across all 73 verticals, see the full methodology overview.