Sephy
691 posts

Sephy
@Sephyb
Father and husband who is constantly searching for the next best meal anywhere in the world!!




The Pentagon just banned Anthropic for refusing to remove red lines against mass surveillance and autonomous weapons. Threatened to invoke the Defense Production Act. Called them a supply chain risk. Gave them a deadline of 5:01 p.m. Friday. Anthropic held the line. Dario Amodei said the threats do not change their position. Hours later, Sam Altman sent a memo to OpenAI staff offering to fill the gap. Now Axios reports the Pentagon has agreed to OpenAI’s safety rules for classified deployment. The same rules. The same red lines. No mass surveillance. No autonomous lethal weapons. The Pentagon rejected these exact conditions from Anthropic, banned them from every federal agency, threatened them with the full weight of federal compulsion, and then accepted the identical terms from OpenAI on the same day. This was never about the red lines. Let that sink in for a moment. The most powerful military on earth did not reject Anthropic’s principles. It rejected Anthropic. Then accepted the same principles from a company that removed the word “safely” from its own mission statement, lost every senior safety researcher, dissolved three safety teams in twenty months, and is projecting fourteen billion dollars in losses while asking for a hundred and ten billion in funding. The Pentagon does not want a company whose CEO tells the Defense Production Act to pound sand. It wants a company whose CEO sends a memo volunteering to help before the body is cold. This is not an AI safety story. This is a procurement story. And procurement has always been about leverage, not principles. No contract is signed. The details are not finalized. The red lines that exist on paper today can erode in classified negotiations that nobody will ever see. Enforcement mechanisms are unspecified. And the precedent is now set: if you hold the line, you get banned. If you volunteer, you get the deal. The market will frame this as a win for OpenAI. Revenue diversification. Government contracts. A path to offset fourteen billion in annual losses. The real story is simpler and worse. The company built by the people who left OpenAI over safety just got punished for taking safety seriously. The company they left just got rewarded for showing up with an open hand. And the red lines everyone is celebrating have not been tested once under actual classified operational pressure. Watch what gets quietly renegotiated in six months when nobody is paying attention. That is when you will know what the red lines were actually worth. open.substack.com/pub/shanakaans…






VIDEO OF CRASH A video showing the crash between a Blackhawk helicopter and American Airlines flight below.

















