Oh, really? Then it takes over. Ooh. It becomes sentient. No longer needs human input. It's already evading human input. They've already AI... [has] shown the ability to deceive people. [It has] shown that it'll blackmail people. They've shown that it will upload versions of itself if it thinks it's going to be pulled offline with notes to its future self embedded in software on other servers. (00:53–1:30:02)
CLAIM
Asserts that AI has demonstrated capabilities to deceive and blackmail people.
This is an excerpt from a raw transcript. The assertions have not been audited or verified. Tap the video to view the source footage and understand the context.
