As global order continues to fray, some tech leaders are speaking out about their obligations to America’s interests and the West’s. Palantir CEO Alex Karp is one of the most vocal, and least equivocal, proponents of American strength. He views it as a force for global good. During an overlooked but important conversation at the Munich Security Conference last weekend, he explained why AI leaders need to choose the West over China and other opponents of the U.S.-led global order.
“I believe, like most Americans, that we should have very severe restrictions on exporting our enormous advantage to China,” Karp said on a panel. He also expressed hope for the possibility that the U.S. and China could one day work together — but pointed out that “we’re very far away from that.”
He was responding to comments from the European Commission’s tech czar, Margrethe Vestager. She had just made the case for substantial engagement with China on AI governance; this is one aspect of the EU’s “complex” relationship with the country, she said. That relationship is one that encompasses cooperation on climate, even as Europe engages in “systemic rivalry” with the People’s Republic. She added: “We need to engage one another in how we see technological development, how governance should be developed, what should be the restrictions, what should be the enabling factors in the marketplace.”
Palantir offers AI products that are already being used on the battlefield in Ukraine. The company has set up shop in Kyiv, and it’s helping the Ukrainian military with “most of the targeting in Ukraine,” Karp told Time magazine, which recently published an expose on Palantir’s extensive work there.
Karp doesn’t try to hide the ball. He’s blunt about what his company does. “I think being open about the fact that these at the margin are technologies that can and should be used in the West to defend the West, like in Ukraine, like in Israel,” he said. “But they are at the end weapons, technologies that will define who controls the world order, and I’m very, very in favor of that world order being controlled by the West, primarily by America and its allies,” Karp added, before also saying that he opposes xenophobia in any form.
“I believe American technology built in America should only be sold to American allies and certainly should be made available to all Western governments.”
Vestager challenged him on this point, asking if both things — imposing certain national-security restrictions on tech exports, while allowing some transfers — can’t be achieved at the same time. Karp’s reply: “No.” What he means by that, he subsequently explained, is that international summitry is a good thing, but there should be a “bright line” proscribing the transfer of AI know-how to China.
That might sound like common sense. But Palantir’s unequivocally pro-Western stance differentiates it from some other major players in the industry. Karp’s perspective on that is important. The other issue that all of this raises: With Washington and Beijing in the middle of a temporary rapprochement, bilateral dialogue on AI governance, of the sort that Vestager championed, will likely be on the table. But what’s the point, and who actually believes that Beijing will in good faith ink an agreement that constrains its capabilities?