There you have it! The Chairman of the Joint Chiefs just admitted at Vanderbilt that the Pentagon…
April 27, 2026 · 0 likes · 0 comments
China Threat Defense Cybersecurity Workforce AI
There you have it! The Chairman of the Joint Chiefs just admitted at Vanderbilt that the Pentagon's procurement system cannot secure AI for autonomous warfare.
His words: "We have to write better contracts."
We have to write better contracts. That is the answer from the most senior military officer in the United States while China is fielding autonomous drones and building AI weapons factories.
Gen. Dan Caine — the guy who left the military to become a venture partner at Shield Capital and Thrive Capital before coming back to make a huge impact — told us the Pentagon's acquisition frameworks were designed for fixed hardware. Not software. Not AI. Not anything that updates more than once a decade.
But here is where it gets truly insane.
While the White House publicly ordered every federal agency to phase out Anthropic tools — designating the company a supply chain risk — the NSA quietly got access to Mythos Preview.
That is the model Anthropic itself said was too dangerous for public release. The model their own red team confirmed can identify and exploit zero-day vulnerabilities in every major operating system and every major browser.
CISA — the agency actually responsible for defending America's networks — does not have access to Mythos.
The NSA does.
JOKES WRITE THEMSELVES.
Ban the company publicly. Use their most dangerous model secretly. The one hand does not know what the other is doing. I have seen this movie before. I was inside that building when the same dysfunction played out on a dozen other programs.
This is not about Anthropic. This is about a system so broken that the Chairman of the Joint Chiefs has to publicly beg for better contracts while the intelligence community does whatever it wants through back channels.
Autonomous weapons are coming. Caine called them a "key and essential part of everything we do." The question is not IF. It is whether we will deploy them with procurement theater or actual security.
Right now it is theater.
How is that acceptable?
Join Episode 2 of my live show — "Inside My AI Kingdom" — May 5 at 1 PM ET. Register on LinkedIn: https://lnkd.in/eZNmtQNW
And if you want the full story of how we got here — REPLACEMENT drops end of July.
His words: "We have to write better contracts."
We have to write better contracts. That is the answer from the most senior military officer in the United States while China is fielding autonomous drones and building AI weapons factories.
Gen. Dan Caine — the guy who left the military to become a venture partner at Shield Capital and Thrive Capital before coming back to make a huge impact — told us the Pentagon's acquisition frameworks were designed for fixed hardware. Not software. Not AI. Not anything that updates more than once a decade.
But here is where it gets truly insane.
While the White House publicly ordered every federal agency to phase out Anthropic tools — designating the company a supply chain risk — the NSA quietly got access to Mythos Preview.
That is the model Anthropic itself said was too dangerous for public release. The model their own red team confirmed can identify and exploit zero-day vulnerabilities in every major operating system and every major browser.
CISA — the agency actually responsible for defending America's networks — does not have access to Mythos.
The NSA does.
JOKES WRITE THEMSELVES.
Ban the company publicly. Use their most dangerous model secretly. The one hand does not know what the other is doing. I have seen this movie before. I was inside that building when the same dysfunction played out on a dozen other programs.
This is not about Anthropic. This is about a system so broken that the Chairman of the Joint Chiefs has to publicly beg for better contracts while the intelligence community does whatever it wants through back channels.
Autonomous weapons are coming. Caine called them a "key and essential part of everything we do." The question is not IF. It is whether we will deploy them with procurement theater or actual security.
Right now it is theater.
How is that acceptable?
Join Episode 2 of my live show — "Inside My AI Kingdom" — May 5 at 1 PM ET. Register on LinkedIn: https://lnkd.in/eZNmtQNW
And if you want the full story of how we got here — REPLACEMENT drops end of July.