The BasePeople

annawang

Titles as the sources gave them. In 1 entry, with 3 checked quotations.

What they said

Newest first. Each quotation was found word for word in the source.

“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.”
Delaying Internal Deployments Until After X Days of Safety TestingAI Alignment Forum, November 11, 2026. Speaking as: Other. The author states the concrete version of the proposal they prefer.
“Today a problem seen in real use can feed into the next model’s training almost straight away; under this proposal, a model will get X more days of training before an issue is detected.”
Delaying Internal Deployments Until After X Days of Safety TestingAI Alignment Forum, November 11, 2026. Speaking as: Other. The author lists this among the drawbacks of the proposal.
“I have a hunch that, if implemented well, this proposal would have a larger impact on the pace of recursive self-improvement (RSI) itself than on the onset of RSI.”
Delaying Internal Deployments Until After X Days of Safety TestingAI Alignment Forum, November 11, 2026. Speaking as: Other. The author gives a tentative view on what the delay would do if used as a speed dial.

In the Base

Every entry they appear in.

  1. EssayAI Alignment ForumWrote it

    Delaying Internal Deployments Until After X Days of Safety Testing

    In a nine-minute post written in a personal capacity, the author annawang proposes that a new AI model not be used inside its developer, beyond safety testing, until a fixed number of days after being filed with an auditor. The post lays out a scenario, open questions, benefits and drawbacks.