The Trump administration briefed major AI labs including OpenAI and Anthropic on its new AI cybersecurity framework this Tuesday. The framework is designed to mitigate risks of AI-driven cyber threats but its details remain classified leaving the public and broader tech community in the dark. The move echoes past administrations’ penchant for selective transparency with big tech while paying lip service to accountability. Wired reports the briefings were part of a closed-door session with no public disclosure timeline.

This is policy capture in real time. By sharing the framework exclusively with a handful of AI labs the White House is signaling that AI governance is a negotiated privilege not a democratic obligation. The lack of transparency undermines trust and sets a precedent where critical safety standards are shaped by the same entities they are meant to regulate. Second order effects include a potential chilling effect on open-source AI development as smaller players lack access to the rules of the game.

Winners are OpenAI Anthropic and other invited labs who gain early influence over regulatory contours. Losers include the public open-source AI developers and smaller startups locked out of the loop. The White House risks eroding its own credibility by appearing to favor incumbents over innovation.

Watch for leaks from the briefed AI labs as engineers and advocates push for transparency. Expect backlash from open-source communities and calls for congressional oversight. If the framework remains secret it could accelerate a bifurcation in AI development between the anointed few and everyone else.