Changing My Mind: Thinking More Seriously About Palantir
Has the binational Big Data company been turned by AUKUS? "Ask the Ukrainians"
I’ve been reading and thinking a lot about Palantir, the data company founded by Peter Thiel and Alex Karp in 2003.
Readers have asked why I’ve resisted being as hard on Palantir as I have been on the Lonsdales or Musks of the world. Isn’t Palantir a product of the binational security state and indeed, hasn’t Peter hired and promoted a lot of strange Likud types, Charles? Didn’t you expose one of those frauds yourself?
True enough but Palantir isn’t Peter Thiel and I’ve been (maybe) a little too hard on Peter in shifting him in the right direction. For that I’m sorry. He’s shown a pragmatic streak which should be encouraged. It was good of him to sit out the 2024 election cycle.
After all as global macro shifts, so too must Palantir and Peter.
That’s what I think is fundamentally missing from the Bear Cave case against Palantir. Is Palantir really worth $30B? I don’t know. Stocks are always worth what people are willing to pay for them. I don’t really believe in operating in the public markets. Too many weird things going on there!
The truth is that Palantir is both a consulting company and a technology company and that it justifiably lacks comparisons. It may well be unique.
Is it possible that Karp is correct about how he’s going to take the market on AI? I don’t know. But can Palantir’s customers really afford not to find out? These large, lumbering companies are most in need of the sort of services that Palantir offers.
While it’s true that Palantir’s AI offerings aren’t really up to snuff (at least yet) the world actually still needs consulting companies. It especially needs consulting companies capable of making difficult decisions like those currently contemplated over the National Health Service. Yes, it’s darkly funny to me that Rishi Sunak — a former Goldman Sachs man — is the one who has to make the call. Isn’t America grand? Alex Karp met up with Prime Minister Rishi Sunak at a Nationals baseball game. You can get a sense of it by listening to the BBC.
Don’t we all know that Starmer’s just marking time before Keir Starmer ascends to No 10 Downing Street? The Financial Times has a rather amazing profile of how Starmer has consolidated power.
The antitrust environment would allow Palantir to acquire these real AI companies—but only if Palantir became a national champion. That’s tricky in an era of AUKUS. My sense is that Palantir must become an AUKUS champion. Harmonizing the regulatory framework between Washington and London is something we will continue to see, especially as both nations contemplate a post-Ukraine war. The United Kingdom will become a sort of laboratory especially as it goes full bore into genetics.
Were I running Palantir, I’d do a few things differently. I’d look for (and acquire) data generating companies, particularly those involved in DNA and biometrics. I’d work with NHS on expanding the crown jewel of the British scientific system — the UK Biobank.
I’d double or triple down on the AUKUS alliance by hiring more from GCHQ and fewer from the finishing schools. Think the clever scholarship boys, not the old Etonians. More Turings and Cricks, please.
In much the same way that Facebook (headed by Nick Clegg) and the Soros Open Society (headed by Mark Malloch Brown) was turned by Brits so, too, does it seem to me as if Palantir is being cleaned up.
I’d acquire (and clean up) companies like Clearview.AI which have a relationship already with law enforcement agencies throughout the United States. The only reason Clearview.AI isn’t even more successful is that few people trust Hoan. (And with good reason as my lawsuit points out.) Should that be impossible, Palantir should build a clone of the firm.
The problems of the National Health Service are legion and will require careful attention. We discussed elsewhere how polygenetic scores might be used to save lives of would be cancer patients.
In my assessment the only real way to drive down the costs of the NHS is through full genome sequencing. The UK Biobank led the world with its offerings but those offerings are not representative of the global population.
At Traitwell, we’ve done a lot of the thinking around how to build up an American Biobank. We increasingly think that what’s needed is an AUKUS Biobank. Palantir would be a great partner, especially in light of its new relationship with Amazon, and their commitment to securing the data.
Here’s how we’re going to build that biobank with a security and private minded focus.
Deceased
One of the fastest ways to build a national biobank is to require that everyone get sequenced upon death. Another way is to minister to the preneeds group. You can see a Labour government under Starmer encouraging all of Britain (and willing members of AUKUS) to get genetically sequenced. “Donate your DNA for faster cures” would be the slogan.
Infants
There’s been a lot of evidence lately that newborns benefit tremendously from genetic sequencing.
What would happen if every newborn's genes were sequenced at birth? That's the question the BabySeq study has been trying to answer for a decade.
Its newest results suggest that genetic information could be used to save lives. And not just the baby's.
Several years ago, the research showed that out of 159 seemingly healthy babies whose genetic information was sequenced at birth, 17 were found to have "actionable" mutations – their genes predicted or increased the likelihood they would fall ill – and knowing that ahead of time could change the course of their lives.
This follow-up, 3 to 5 years later, shows the information was useful, both for the babies and three of their mothers. Most genetic diseases are inherited from either parent or a combination of the two, so the baby's mutation indicated one parent's genes carried a similar disease risk.
Nebula offers sequencing for $600 and falling. Nebula can do about 185,000 samples a year. What would a world that sequenced a million people a year look like?
Migrants
The CBP One app would require DNA ultimately for all the migrants coming into the U.S. Right now they collect DNA using the CODIS tests which don’t give nearly enough information to be useful. There will be increased migration due to climate change. Best to get a hand on processing the migrants quickly.
Prisoners
The ACLU is suing so that every prisoner can get access to their DNA. Palantir should support this effort as well as part of their commitment to privacy and civil liberties.
Soldiers
As the military becomes smaller knowing what each soldier is fully equipped for is key. This’ll help reduce the costs too.
Ultimately Palantir should help build a world class lab in the Northern Virginia area. Once Illumina is cleaned up there’s a deal that can be constructed there with their sequencing machines. That’ll happen by 2025.
All of this can be discussed but it’s nice to see Palantir back in the mix.
It’s so refreshing to see America and Britain working together again.
The real master — Francis Galton — is the path to follow, Palantir. Let’s go!