Post #4769638 · Elektrine

2026-09-11 09:22 UTC

@Captain_Jack_Sparrow@mastodon.world @_elena@mastodon.social @timnitGebru@dair-community.social @emilymbender@dair-community.social @alexhanna @DAIR@dair-community.social @cwebber@social.coop @jaredwhite@indieweb.social @tante@tldr.nettime.org I think it is a scam, but also dangerous - however, the danger is not inherent in the AI tools (IMO). It is inherent in the entire software infrastructure we have. This was a massive risk even before AI tools. The difference is that AI can really easily exploit this, much easier than humans can.

Replies (2)

@Captain_Jack_Sparrow@mastodon.world 2026-09-11 10:04

@SteveClough@metalhead.club @_elena@mastodon.social @timnitGebru@dair-community.social @emilymbender@dair-community.social @alexhanna @DAIR@dair-community.social @cwebber@social.coop @jaredwhite@indieweb.social @tante@tldr.nettime.org I agree with you, and if quantum computing becomes a thing, then all our passwords will be useless.

Open ##4769639
@MySideIsHumanity@myprivate.social 2026-09-18 10:21

@SteveClough@metalhead.club @Captain_Jack_Sparrow@mastodon.world @_elena@mastodon.social @timnitGebru@dair-community.social @emilymbender@dair-community.social @alexhanna @DAIR@dair-community.social @cwebber@social.coop @jaredwhite@indieweb.social @tante@tldr.nettime.org Thank you for raising this point, and I agree from a slightly different tack: It’s not ~inherently~ dangerous/unregulated, a la gunpowder per se. But the “robot defense” needs to be explicitly forbidden by law. Corporations can’t be allowed to perpetrate profitable crimes, and argue their innocence or stupidity (“the ai did it!”) a la DUI in decades past.

Open ##4769640