The transition to an engineering discipline in drug discovery, analogous to aeronautics, means using powerful in silico models to get much closer to a final product before physical testing. This reduces reliance on iterative, expensive, and time-consuming wet lab experiments.

The next leap in biotech moves beyond applying AI to existing data. CZI pioneers a model where 'frontier biology' and 'frontier AI' are developed in tandem. Experiments are now designed specifically to generate novel data that will ground and improve future AI models, creating a virtuous feedback loop.

While AI promises to design therapeutics computationally, it doesn't eliminate the need for physical lab work. Even if future models require no training data, their predicted outputs must be experimentally validated. This ensures a continuous, inescapable cycle where high-throughput data generation remains critical for progress.

The most significant breakthroughs will no longer come from traditional wet lab experiments alone. Instead, progress will be driven by the smarter application of AI and simulations, with future bioreactors being as much digital as they are physical.

AI's primary value in early-stage drug discovery is not eliminating experimental validation, but drastically compressing the ideation-to-testing cycle. It reduces the in-silico (computer-based) validation of ideas from a multi-month process to a matter of days, massively accelerating the pace of research.

A key strategy for labs like Anthropic is automating AI research itself. By building models that can perform the tasks of AI researchers, they aim to create a feedback loop that dramatically accelerates the pace of innovation.

Instead of generating data for human analysis, Mark Zuckerberg advocates a new approach: scientists should prioritize creating novel tools and experiments specifically to generate data that will train and improve AI models. The goal shifts from direct human insight to creating smarter AI that makes novel discoveries.

The ultimate goal isn't just modeling specific systems (like protein folding), but automating the entire scientific method. This involves AI generating hypotheses, choosing experiments, analyzing results, and updating a 'world model' of a domain, creating a continuous loop of discovery.

Contrary to the idea that AI will make physical experiments obsolete, its real power is predictive. AI can virtually iterate through many potential experiments to identify which ones are most likely to succeed, thus optimizing resource allocation and drastically reducing failure rates in the lab.