Artificial Intelligence
Out of context: Reply #2665
- Started
- Last post
- 2,691 Responses
- NBQ003
The Trump admin is still bitter because Anthropic refused its LLM's for warfare.
From reddit:
TLDR; Every time Anthropic pulls ahead, the US Government finds a "security" reason to hit it - and OpenAI is the one left standing.
Nov 2025 - OpenAI locks in ~$1.4T in compute promises against ~$13B revenue. Can't afford to lose.
Feb 27, 2026 - Trump orders all federal agencies to drop Anthropic and the Pentagon labels them a "supply chain risk"
Feb 27, 2026 - OpenAI signs its own Pentagon deal the same day, hours later.
May 22, 2026 - OpenAI confidentially files for IPO targeting $1T+ - while valued at $852B and losing roughly $1.22 for every $1 of revenue (~$27B cash burn for 2026).
Mar 24, 2026 - Federal court grants Anthropic an injunction, ruling the designation was punitive retaliation, not a security call.
May 28, 2026 - Anthropic's $65B Series H closes at a $965B valuation, passing OpenAI's $852B. It's now the most valuable AI lab in the world.
Jun 1, 2026 - Anthropic confidentially files for IPO, expected to target $1T+. (OpenAI is racing to list this year too, carrying ~$1.4T in compute commitments against ~$13B in revenue.)
Jun 9, 2026 - Anthropic ships Fable 5 + Mythos 5, the most capable models it's ever released.
Jun 12, 2026 - Commerce Sec Lutnick pulls BOTH for all users citing "national security." Letter gives zero specifics. Anthropic says the flagged "jailbreak" is a capability OpenAI's GPT-5.5 already has - and GPT-5.5 stays up.
Also Lutnick - under a congressional ethics request over his family's AI data-center interests.
Summary: OpenAI has a ~$1.4T hole of promises, racing to a $1T IPO (it's losing money to justify). OpenAI keeps getting 'bailed out' by "security" rulings that land on its only real competitor - the moment that Anthropic pulls ahead. A judge already called February retaliation. This is the same move with a fresh label.