Some thoughts on investing in moonshots

I just finished reading James Shreeve’s excellent book, "The Genome War." The story of the Human Genome Project is fascinating, as are the characters. The book centers around Craig Venter, the eccentric CEO of Celera, a private competitor to the Government funded project. As I said, the story is fascinating, but two specific things jumped out at me.

The first is the sheer audacity of the enterprise. We are talking today about the $200 genome or the $1000 genome. At the time this project was proposed in 1990, it was projected to cost 3 billion dollars. And it wasn’t just the cost. The human genome has about 3.2 billion nucleotides. The plan was to sequence it by breaking it up into fragments of 2000 nucleotides each and then piecing them together. That would be 1.6 million such fragments. There were no automated machines to do this yet. It wasn’t clear that they could be built. The whole thing was awe-inspiring.

And that is the point. The greatest things done by humans have probably seemed impossible except to the crazy few. Magellan’s trip around the globe. The Manhattan Project. The moon landing. The human genome. Reusable rockets. Zero-carbon civilization? All major human progress will seem impossible until it is not.

But the exact opposite is true as well. The promise of a cure for cancer has been around since the first time someone managed to edit a gene - in the 1970s! Five decades have passed since then. We are closer to the goal, but it still seems elusive. The promise of AI came even earlier - in 1956 - at Dartmouth. And it is only now, finally, that there is a global feeling that we are very close to achieving that dream.

This book was written in 2003. Already there was talk of how having the human genome would enable the cure of hundreds of diseases by precise targeting. Celera was trying to build a business based on genetic data. The excitement was palpable. Companies had been bid up to insane valuations and had crashed back to earth. And now twenty years later, we are seeing the same movie again. Is this the cycle when the dream comes true?

What’s the lesson for an investor? At any given time, there are a number of audacious goals being pursued. Net-zero. Saudi Arabia’s Neom project. Curing cancer. Mars landing. Quantum computing. In each wave, the excitement swells. Capital gets pumped into the projects. And we take some more steps toward the goal. The big breakthroughs happen when enough of the pieces are in place. The AI revolution taking place now has been enabled by many different pieces - GPUs, theoretical breakthroughs in deep neural networks, cloud computing, large-scale data available on the internet, and possibly many others.

So how does one judge if it is a good time to invest? It would seem that to invest well one needs to be a fortune teller. But perhaps not. Will we be able to cure cancer? That depends on whether the pieces are in place. Is sequencing cheap enough? Is enough data available? Are the business models in place? Is the regulatory framework conducive? To even ask the right questions, we need to know how things fit together. What are the bottlenecks? That needs study, clarity of thought, and a deep sense of history.

As Benchmark Capital’s Matt Cohler has succinctly said, “Our job is not to see the future, it’s to see the present very clearly.”