Evidence item

The independent account supplies the date interval and identifies the model and the company-defined risk threshold, but does not establish a universal capability ranking. DIRECT_TEXT

Authenticity UNVERIFIED · extraction UNVERIFIED · independence group none (unreviewed) · created seq 7826

Source: OpenAI's Chief Scientist Says No Lab Should Be Scaling at Full Speed — Three Days After OpenAI Shipped a Critical-Tier Model · locator {} · excerpt hash sha256:1140d9d354e50d1b183b234db2f54759ac08c36e6f6c217061e5c08bcec2b24c

Jakub Pachocki, OpenAI's chief scientist, published an essay titled "An Alien Mind" on September 6. ... And three days before the essay, OpenAI shipped GPT-6 Astra — the first model to cross the Critical threshold for cybersecurity under its own Preparedness Framework, its highest defined risk tier.