BioNTech at $90: why the promise of AI in oncology does not yet fit into a sales multiple

There are two legitimate ways to look at BioNTech, and the appeal of Simply Wall St's analysis is that it places them side by side without picking one. The first is qualitative: a company that co-developed one of the mRNA COVID-19 vaccines with Pfizer and is now reorienting that platform toward personalized oncology, with deep-learning-assisted neoantigen design as its spearhead. The second is arithmetic: a price-to-sales ratio of 7.1 times, above the level considered fair (5x) and above the peer average (6.2x). Between those two readings lies a gap of 82% relative to the fair value of $499.94 defended by the most widely followed community narrative.

That figure should be read with a cool head. A fair value derived from "aggressive revenue expansion" and "high margins" is not a fact, it is a scenario: it describes what the company would be worth if everything went as expected. The sales multiple, by contrast, describes what the market is paying today. The tension between the two does not reveal who is right, but rather the risk taken on by betting on the story versus the present-day snapshot.

The article itself identifies where the bullish thesis could break, and the points are the right ones: that the clinical trials in oncology disappoint, and that AI-driven drug design fails to translate into commercially viable therapies within the timeframes the market is pricing in. Here lies the nuance that distinguishes AI in health from AI in software. In the sector—where Recursion, Exscientia and Insilico Medicine also operate—the promise of shortening development cycles has been on the table for years, but the late clinical phase remains a bottleneck that no model has resolved systematically. Biology sets its own pace, and that pace is not accelerated by more compute.

That said, BioNTech starts from a position few long-term bets can show: a considerable cash pile inherited from the COVID era that, according to the article's profile, gives it room to fund its R&D without diluting shareholders in the short term. That long runway is precisely what allows it to wait for the science to mature without the pressure of raising capital at every bump.

The calm reading, then, is not to choose between $90 and $499, but to understand what separates one figure from the other: clinical execution and time. The convergence of generative AI, computational biology and mRNA platforms is one of the most active investment fronts of 2025-2026, and the deals struck by Roche, Novartis and AstraZeneca with AI startups validate the direction. Validating the direction, however, is not guaranteeing the destination. For an investor, the useful question is not whether the technology is transformative—it probably is—but how much of that future they are willing to pay for in advance.