Fri 2 Oct 2026

UTC Live

Super Intelligence
News and research
Every item sourced

SUPENCE

Superintelligence latest news and info

Tips and corrections
Write to Supence
Contact form

Source Alignment Forum · Published · Open the original ↗

Lab notes

Delaying Internal Deployments Until After X Days of Safety Testing

This post reflects my own takes, and not that of my employer's. This is a simple idea: before a new model can be deployed internally, you must first test it for a minimum, fixed period. You can adjust various axes of this proposal: Duration of the wait: in calendar time, or until you fulfill some condition. What it blocks: external deployment, internal deployment, or the training of future checkpoints. When the fixed period starts: from the first checkpoint, the final checkpoint, or any checkpoint that is “sufficiently similar / representative”…

This is a simple idea: before a new model can be deployed internally, you must first test it for a minimum, fixed period.

One concrete target that I like is: “No checkpoint may be used internally for purposes beyond safety testing, until at least X calendar days after its hash is filed with an auditor.”

The opening of the post, quoted unchanged from Alignment Forum. The full text continues at the source.

Continue reading at Alignment Forum ↗

More on Supence Newest first

  1. United Airlines Is Banning AI Bots from Booking Flights

    Capability · The Wall Street Journal

  2. Can markets find the AI cash?

    Markets · Financial Times