In the event of shutting down GPT-4, the lab would run evaluations to test for dangerous capabilities such as deception, code manipulation, and independent money-making. If the model proves to be too smart and dangerous, the lab would hit the stop button. However, there is no clear plan for what would actually happen. To address this, it is suggested that pause workshops be held to practice how to handle a pause situation, including informing the board, investors, and employees.

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode