By Barry Buck (with Claude) – Somewhere along the line, tech companies became wartime targets. Iran’s Revolutionary Guards published a list of 18 US tech firms – Apple, Google, Meta, Nvidia, Palantir – and threatened to destroy their “relevant units” if more Iranian leaders are killed, advising employees to immediately leave their workplaces. Happy April 1st from Tehran.
Except it’s not a joke. Palantir’s chief technology officer bragged on American television that AI-powered planning enabled 2 000 strikes in 48 hours during the Iran conflict – with Anthropic’s models reportedly under the hood.
The same Anthropic that just won a preliminary injunction against the Department of War after being designated a supply-chain risk for insisting on human-in-the-loop safeguards. So the company whose AI helped disable Iranian air defences is simultaneously being punished by its own government for having ethics.
Game theory doesn’t get more absurd than this.
Anthropic’s own April Fools gift to cybersecurity? A leaked blog post accidentally revealed Claude Mythos – described internally as a “step change” with “unprecedented cybersecurity risks”. Cybersecurity stocks slumped.
Anthropic is privately warning officials that Mythos makes large-scale cyberattacks significantly more likely this year. The secret sauce just got hotter.
Cold war behaviours are back. China barred the CEO and chief scientist of Manus – the AI agent startup Meta acquired for $2-billion – from leaving the country. The founders had relocated to Singapore before the sale, apparently believing their own simulations.
They forgot about the social credit system. I’m guessing they ran their risk analysis on Ollama instead of Anthropic – rookie error. And NASA is sending astronauts around the moon today on Artemis II.
Without Kubrick this time, though the original “first photo of Earth” was also conveniently released on April 1st – still a better gag than their colleague’s slice of chorizo masquerading as a distant planet.
The lines between war and tech have never been blurrier. AI enabling precision strikes. The same models banned by the governments they protect. Leaked capabilities tanking markets. Founders trapped by exit bans. Rockets heading moonward. It’s less a news cycle and more a season finale.
Which brings me to my actual question: remember when game shows had a “phone a friend” lifeline? Anthropic just shipped Claude Code Channels – letting you message Claude over Telegram and Discord like a personal assistant. Has anyone updated the game show format? Because Claude is one of my most reliable friends at this point. And I’d bet my final answer on it.
Barry Buck is the chief technology officer of Saucecode and Roboteur architect