Open AI: A Masterclass in Branding Failure
Share
How OpenAI violated nearly every law of meaning perception and power and handed Anthropic the throne
(I know, I know I am rude…)
It was late February 2025 when Anthropic stood firm, resisting Pentagon demands for 48 hours to remove safety guardrails from Claude AI. They refused to enable autonomous weapons or mass surveillance. In response, the Defence Department blacklisted them, citing concerns over supply chain risks that could lead to a $5 billion loss.
Then, OpenAI took action. Within a few hours, they announced that they had secured their own artificial intelligence contract with the Pentagon.
“We shouldn’t have rushed to get this out on Friday,” Sam Altman admitted later. “It just looked opportunistic and sloppy.”
(No Sam. It looked like Law 48 in reverse.)
They totally freaked out and abandoned their own playbook. Instead of managing the crisis, they caused it! But hey, at least everyone was talking about ChatGPT—reviews skyrocketed 775% that weekend. So much for just being “sloppy.”
What OpenAI actually demonstrated was Law 3 inverted. Appear Simple Think Complex. They appeared complex, rushed, reactive, and desperate while thinking simply. Short-term revenue over long-term meaning. The Google homepage is blank because it runs the world’s most advanced algorithm. OpenAI’s announcement lacked meaning because it was driven by pure opportunism. (Genius move, really.)
The Military AI Contract Nobody Wanted: Law 16 in Action
Anthropic built its brand on Law 16. Let Your Values Dictate Your Boundaries. Their Constitutional AI wasn’t marketing. It was the landmines you’re willing to step on. When the Trump administration demanded all lawful uses, Dario Amodei refused. “We cannot in good conscience accede to their request.”
The punishment was swift. Designate Anthropic as a supply chain risk. Tell every defence contractor to drop them. Destroy an AI startup for having ethics. Because nothing says free market capitalism like punishing companies for having principles.
This was supposed to be a death sentence. Instead, Law 1 manifested. Own a Meaning Not a Market. Anthropic didn’t own the military AI market. They owned the meaning of AI safety first. And that meaning, as you wrote, holds when markets move. By Saturday, Claude became the number one downloaded AI app in America. Consumers were choosing meaning over market presence. (Revolutionary concept, apparently.)
Law 26 clarifies what happened next. Create Enemies When Necessary. Anthropic didn’t seek the Pentagon as an enemy. But by refusing to compromise, they sharpened their story. The heat from the "villain administration" just made the hero's principled AI look even better. A major contrast, just like Law 8 talks about. In a sea of pastel tech startups choosing profit, Anthropic chose blood red principles. (How dramatic.)
