We’ve now seen multiple trends come to fruition at the intersection of bio and technology over the past decade: A Moore’s Law for bio, thanks to computation; machine learning and AI transforming many areas of bio pharma and healthcare; the ability to not just “read”, but “write”, to bio, including CRISPR (even in just a decade). We’re also seeing the rapid unbundling of care delivery as well now, driven by “the great unlock” — which includes the unbundling of the hospital (into decentralized communities, virtually, and more) and the movement towards value- or outcome-based care.
Driving this revolution has been a new breed and wave of founders and startups that merge the worlds of technology and bio — importantly, not just the old world of biotech (or a narrow definition of tech in bio as only “digital health”), but something much broader, bigger, and blending both worlds. In short, biology — enabled by technology — is eating the world. This has not only changed how we diagnose, treat, and manage disease, but has been changing the way we access, pay for, and deliver care in the healthcare system. It is now entering into manufacturing, food, and several other industries as well. Bio is becoming a part of everything.
This new era of industrialized bio — enabled by AI as well as an ongoing, foundational shift in biology from empirical science to more engineered approaches — will be the next industrial revolution in human history. And propelling it forward is an enormous new driving force, the novel coronavirus SARS-CoV-2, its ever-evolving strains, and the resulting COVID-19 disease pandemic and response — which I believe is analogous to our generation’s World War II (WW2). In other words: a massive global upheaval, but that later led to unprecedented innovation and significant new players.
As a result, we will now see the emergence of bio’s version of GAFA — playing off the “Google Amazon Facebook Apple” of the leading companies in computing, social, mobile — but for bio. And with it, a post-WW2/ post-Covid “Industrial Bio Complex”.
The World War II moment in tech x bio
Over half a century ago, we saw the modern industrial revolution decades after the war. It began with advances that came out during or immediately after the war, in healthcare — like flu vaccines, commercial production of penicillin, and blood transfusion — to technology advances like jet engines, radar, microwaves, and electronic computing (most notably with ENIAC, one of the world’s first general-purpose computers). But it was innovations in mass manufacturing and automating production — of household items, cars, etc. — that happened decades after, in the 1970s, that led to the most recent industrial revolution with later advances in semiconductors, electronics, computing, AI, and more.
As with WW2, COVID has been a brutal wake-up call and shock to the global system. The mass casualties and horrors of this disease, as well as our ability to adapt and improve our response to it, will spur — has already spurred — a new wave of innovation, renewed energy, and vigor in the biopharma and healthcare industries. The world now sees both the why and the how behind the promise of engineering in healthcare; just take for instance the speed at which and way we developed the mRNA vaccines.
This is why I believe the equivalent tech revolution that follows our generation’s world war, post-Covid will be the creation of an Industrial Bio Complex — because the industrialization of bio brings new scale, and with it, new treatments, healthier lifestyles, new manufacturing and construction that reduces the effect of climate change, and more. Bio thus becomes the new manufacturing revolution, where everyone pours their attention, investment, education, and efforts: whether for safety and health, national security, or greater prosperity.
Artificial intelligence (AI) will continue to play a transformative role here. As with previous revolutions, this technology platform shift will impact industries differently. Notably, international scholar Carlota Perez has illustrated how the technologies of the next industrial revolutions are not necessarily an aid to the previous; for example, the tech revolution of the 1970s did not majorly impact the industrialization of 1940s and 1950s goods. Similarly, AI ironically will not have its greatest impact on the tech revolution. For areas where “compute” already dominates — that are already riding Moore’s Law — AI can be overhyped; and as Casado and Bornstein explain, AI does not necessarily make a massive contribution to those business models. [It is tempting to label AI itself as an industrial revolution, but as Perez further notes, the technologies themselves are not revolutions. Rather, industrial revolutions reflect the combination of multiple, transformative technologies into a specific, large potential market.]
In the bio and healthcare market, which is largely dominated by services, we would expect to see potentially massive gains from AI — at least to the extent that AI can transform services into “compute”. AI is not always as useful in enterprise because it is slower than the status quo, more expensive in terms of unit economics, etc. But in bio, AI can be extremely helpful: It helps turn the things that used to be expensive, human-labor intensive, less efficient, and less accessible, into less costly, more efficient, and even more effective “compute”.
Once that happens, technology can penetrate into the industries that previously did not benefit from the IT revolution. In fact, economists and innovators have long wondered why we haven’t seen the kinds of massive gains we’ve seen in other industries, in healthcare — aka Baumol’s Cost Disease — because when technology successfully penetrates industries, it transforms formerly expensive services into much more affordable goods. (And frees up human labor to work on more meaningful work, which is where it could help in healthcare. In other words, technology can make human healthcare more human.)
We are poised on this cusp today. Until now, healthcare and biotechnology have been heavily dominated by services — provided by expertly trained scientists and doctors — that algorithms could not replace, let alone add enough value, to make sense for companies to adopt them. But now, we are at the very beginning of a revolution where AI is industrializing biopharma and healthcare, and it is being applied to everything from drug design and diagnostics to healthcare delivery and back-office functions. [As for concerns or challenges that often come up with discussions of applying AI in bio, I address the “black box” of AI in healthcare here; and address what’s needed for us to get smart (vs. “dumb”) AI in bio, here.]
In all these areas (and others yet to be seen), the industrialization of biopharma and healthcare can go a long way. Instead of a person training in an apprenticeship-like way to do something, for instance, the machine learns those things. We can make copies trivially, which lets us scale expertise in the same manner as we spin up cloud-based servers: immediately, cheaply (compared to expert labor), and at immense scale. And this is not limited purely to compute: Automation and robotics similarly enables scale, as well as further enabling true reproducibility of biology in a way not previously possible — thus addressing one of the greatest weaknesses of “pre-industrialized” bio.
Bio can now scale, massively.
Where and how do we go from here?
Drawing on lessons from the last two decades in *tech*, what would we expect to see for the next two decades in *tech x bio*? What happens as we move even more so into engineering biology, and what are the consequences of two mega trends — the industrialization of bio, plus the post-WW/ post-Covid effect — coming together?
Consider for instance the last decade of mobile, and the cycles of how innovation happens in technology. Referencing Carlota Perez’ influential work again: Mobile, like many technologies before it, had an “installation” phase that preceded a “deployment” phase. Perez’ concepts are summarized well here by Jerry Neumann, but basically, in the installation phase, capital flows to building the infrastructure that technological revolution needs; this installation phase also requires more push than pull, which is where go-to-market, partnerships, and other business development especially matters for bio startups here. Then, in the deployment phase of a particular technological revolution — which mobile reached after we passed 2.5+ billion smartphones on earth — companies often move towards expanding and consolidating their existing markets, not just creating them. In this phase, companies begin competing even more so on cost, usability, and scale; for mobile, the shift led to new questions, such as “what can we build as we stand on the shoulders of giants”? (The giants in that case being companies like Apple and Google.)
I expect tech x bio to follow a similar path. Right now, we are at the juncture of realizing we need the ability to engineer bio, and a fuller ability to engineer it (i.e., we’re still in the installation phase). In the tech industry, the analog period for the web led to the creation of massive companies like Amazon and Google. Given the power of the combined trends — and the scale of the challenges and massive market of healthcare — we should expect to similarly see the rise of a few potentially trillion-dollar companies at scale: the equivalent of a Bio GAFA, finally.
For tech x bio founders, this means their ambitions can be bigger and more attainable than ever, if they are able to execute. For the industry, it means ways to address the systemic and structural problems of healthcare, which are currently fragmented or consolidated in all the wrong ways (not to mention entrenched in legacy silos and structures that technology can re-route around). Most importantly, for all of us, the gains of the Industrial Bio Complex should lead to healthier, longer, and richer lives through greatly improved care and access, at lower cost.
Although it’s still early, there are a few tell-tale signs at this early-stage of what an emerging, future Bio GAFA company would look like:
Full stack companies
The rise of full stack companies — which build a complete, end-to-end product or service — can bypass incumbents and other competitors, if executed well. An analogy that is often used on the consumer side is ridesharing, where many companies had previously tried to build and sell software directly to the taxi/limo industry, but those companies were not set up to assess let alone absorb such software. It was only when companies like Lyft and Uber came along and asked what the industry could look like if it were rebuilt from scratch that using technology for ridesharing finally arrived.
Similarly, incumbents in healthcare are not prepared to absorb advanced technology. Merely adding AI or other technology on top of an existing organization will not have as great an impact as redesigning the entire organization from the ground up to utilize that technology, natively, as you can only do in a startup.
In healthcare, there are many business models for this, but one framework for scalable business models to use here is “compete or connect”, which Vineeta Agarwala has written about. Bio companies that “compete” aim to become full-stack competitors to existing players in the big three bio-enterprise categories (life sciences, provider, or payor) — as opposed to selling into those industries (“connect” to customers via others). This compete-or-connect framework could also help founders and investors assess which bio companies, out of thousands of startups, are likely to have the greatest impact and scale, and which business models will endure the test of time.
Government as buyer, not builder
The Manhattan Project, in its early forms, was formed by the U.S. president and government right before and during World War II, combining — and coordinating — various research and technology efforts to help fight the war. Vannevar Bush, the Director of the then-Office of Scientific Research & Development that administered the Manhattan Project early on, later also wrote the famous “Science, the Endless Frontier” report — which led to the creation of the National Science Foundation. However, I do not see a Manhattan Project for healthcare (or for the environment) emerging from the U.S. political system today, which is not structurally set up for this scale of coordination and technological innovation. Take just for one example how we struggled with the scale and coordination required for implementing testing, as compared to the private sector response, technological capabilities elsewhere, and who has the ability to innovate faster/better.
But government does have an enormously consequential role as a buyer. In particular, the Centers for Medicare & Medicaid Services (CMS), part of the Department of Health and Human Services (HHS), have long been a locus and driver of innovation for healthcare — whether it’s through the Medicare and Medicaid insurance programs; policies incentivizing data sharing, pricing, and transparency; utilizing alternative site-of-care; and adopting value-based payment schemes. CMS tends to be the tip of the spear to the broader payor market, especially in terms of reimbursement — when they implement new reimbursement policies and value-based payment programs, the commercial payors tend to follow suit. Julie Yoo often observes that government-sponsored healthcare payment programs are “the tail that wags the dog” on not just the ~$1.5 trillion spent under Medicare and Medicaid every year, but also the $1.2 trillion spent in the private health insurance market.
As such, in the coming Industrial Bio Complex, the winners will inevitably embrace working with federal and local government programs to tap into the massive payment rails, once they get to population-level scale. Current category leaders are already doing this by leveraging their platforms — which are lower cost and more nimble — to be the technology “execution arm” for many of CMS’s new value-based care initiatives, focusing on senior-citizen and low-income patient populations. It often plays out, Yoo observes, in the form of advanced primary care models or full-stack tech-enabled carriers.
Bridges and builds from both sides
While the ambitions of the next Bio GAFA companies will be huge, and necessarily visionary — the companies that achieve this will be more grounded in reality. To succeed, they will recognize the challenges of applying bio and technology, that in practice very much still remain.
A drug company that assumes biology is “easy” or as predictable as in other areas of tech will fail in its attempts to engineer biology, even with the most powerful AI. A healthcare delivery company that likewise underestimates the complexity and challenges of the U.S. healthcare system will also fail. The successful companies will not only bring a healthy dose of reality, but bring together experts from both sides — tech and bio — and from the ground up, in building their companies.
Along these lines, tech, biopharma, and healthcare are often treated in startup-land as separate worlds, with different sets of investors for each. But we are already seeing these lines blur: Therapeutics take on more of healthcare delivery, as we’ve seen in engineered cell therapies. Drug companies go directly to payors. And diagnostic companies seek to change the very nature of care.
Assets → Platforms → Channels
In biopharma, we are already seeing companies go from focusing on assets to platforms (as Jorge Conde outlines) — thanks to AI, automation, and other engineering that turns bespoke craftsmanship into industrialized drug development. However, the trajectory can go even further: these companies can also become channels. This is another potential indicator of the companies that will realize Bio GAFA.
Traditional biotech companies are often so-called single-asset companies — developing, testing, and commercializing a single therapeutic coming from discovery — in what was a somewhat serendipitous search for some drug that actually worked in the complex context of biology. Now, new age platform companies can engineer subsequent therapeutics from earlier designs, building on knowledge in an iterative way, given the ability to derive and replicate via technology (often very quickly). AI in drug discovery companies can use industrialized processes — automating biology and chemistry experiments that feed into advanced machine learning technologies — to turn what was formerly manual and had a low hit-rate into something that is more repeatable, predictable, and mass-industralizable. And even in its misses, these technologies allow us to learn even from failures, improving (in general) through successive iterations; we are already starting to see this in CRISPR and CAR T.
As bio platforms become more productive than ever before (able to deliver on the promise of multiple new therapeutics), the bottleneck shifts from discovery to development. So early-stage startups with finite resources will need to prioritize and pace the diseases their (otherwise broadly applicable) platform can address, and then interrogate and advance the promising drug candidates through clinical development hurdles (either in-house or via partnerships). But the productive platform that also figures out distribution for itself can now become the channel for many others, further leveraging what they’ve built: They can sell multiple products into the market, much like Amazon used its marketplace and commercial infrastructure to go beyond selling books to selling, well, everything.
And, naturally, this won’t be limited to biopharma. Technology-enabled approaches to Medicare — assembling patients, providers, and payors — creates channels for other applications (such as telemedicine and novel diagnostics), all aligned by value-based reimbursement rails. Companies can now more effectively help patients by becoming a single, direct point of contact with patients, enabling a broad range of technologies that can decrease costs and impact patient lives.
Building the bio economy
There’s greater potential than ever to engineer biology into huge markets beyond just pharma and health — manufacturing, construction, durable goods, etc. — especially in ways that combat, rather than exacerbate, climate change. There’s growing consumer demand for all this, but to date, there haven’t been many strong places (in bio) to satisfy it.
Furthermore, such technologies applied to bio could raise the prosperity of many more countries and workers around the world. For instance, just by replacing the costly, inaccessible logistics cold chain through which we get our fruits and vegetables with applied material-science advances, entire countries could advance — much like mobile phones enabled countries to leapfrog into the modern era. We’re seeing similar advances in the engineering of food itself for health and global sustainability.
Another huge potential opportunity along these lines is the ability to replace petroleum-based plastics infrastructure with biomaterials, including even those that can sequester carbon from the air, allowing near zero (or even negative, given cap-and-trade) cost of goods. While these advances have not fully arrived into consumer’s hands yet, the roadmap for building them is clear, as is the power of consumers who want to transform the status quo. These forces may finally lead us into a more bio-based, than petroleum-based, economy.
With bio, an unprecedented era of innovation awaits.
* * *
The Covid pandemic, even when it ends, will have lasting consequences: Not only in accelerating several trends (we wrote about 16) that transform biology, medicine, and healthcare, but in other industries, too. While I believe it is our generation’s world war, whether or not you agree that’s the correct analogy, there are undeniable parallels to the pre- vs. post-world war waves of technology innovation here.
We haven’t yet recovered from let alone ended this Covid war, but when we do, we’ll usher in an era of bio-based development that should lead to massive, tech-like Bio GAFA companies. What are we going to build? And what are we going to learn from the lessons of the past?
Given then-President Eisenhower’s warning of the Military Industrial Complex post World War II, the more obvious WW2 analog here could be a Bio Industrial Complex — a ramping up of government’s interaction with Bio in the way we saw with Defense post WW2 (with both positive and negative connotations, because it’s certainly true that any powerful technology has the potential for both good and harm). But my point here is that is not something government can create — even if it could, we would likely end up with the equivalent of a $700B/year bio budget 20 years from now, buying tech nobody wants or needs, while creating more problems than solving them.
The difference between the Bio Industrial Complex, then, and the Industrial Bio Complex — which leads with the industrialization of bio and comes to fruition with Bio GAFA — rewards the best of the 1000 experiments that can bloom in the marketplace of startups. These companies can’t be made, but are born. And the only way we can solve or even materially address the enormously challenging and consequential problems of our time is through such forces and scale. In the end, choosing to harness these forces for good is what will allow us to build the Industrial Bio Complex — something that takes the benefit of scale, for a better world.