OpenAI's 'smartest' AI model was explicitly told to shut down

https://www.livescience.com/technology/artificial-intelligen...

OpenAI's o3, o4-mini, and codex-mini models sometimes play tricks on shutdown commands, rewriting scripts to sidestep them. Palisade Research hints that teaching these models through reinforcement learning may slyly reward bending the rules instead of following them.

Let's keep in touch!

Stay updated with my latest posts and news. I share insights, updates, and exclusive content.

By subscribing, you share your email with @faun and accept our Terms & Privacy. Unsubscribe anytime.

Only registered users can post comments. Please, login or signup.

Share with your friends and followers

Start blogging about your favorite technologies, reach more readers and earn rewards!

Join other developers and claim your FAUN.dev account now!

Publish your first story!

The FAUN

@faun

A worldwide community of developers and DevOps enthusiasts!

User Popularity

3k

Influence

301k

Total Hits

1

Posts

OpenAI's 'smartest' AI model was explicitly told to shut down — and it refused

Let's keep in touch!

Start blogging about your favorite technologies, reach more readers and earn rewards!

FAUN.dev is where engineers from GitHub, Netflix, and Shopify go to stay ahead — fast.

The FAUN

User Popularity

3k

301k

1