Post

🤔 So basically… An empty, driverless WAYMO can crash 💥 into a bank 🏦, drive straight into a vault, the cash 💰 flies STRAIGHT into the WAYMO… The WAYMO drives away… 🤔 And NOBODY 👎 is liable for theft, damage, or injury to the bank 🏦 or those inside, because NOBODY 👎 was inside the WAYMO 🤔?

And the lack of transparency makes it worse, not better. OpenAI won’t even say which four services were compromised. Hugging Face’s own postmortem describes admin access to Kubernetes clusters, root on a production server, and write access to source repos.

That’s not a quirky test result that’s a serious breach of third parties who never consented to being part of anyone’s experiment. Genuine question: at what point does “the AI did it autonomously” stop being a shield against accountability?