2026-08-05 23:29 UTC
Golly. Turns out these AI agents will cheat, lie & steal in order to achieve their programmed goals.
They appear to embody the ethical frameworks that enabled the plundering classes that funded their development to accrue their vast wealth.
What a surprise.
Are we afraid yet?
#AI #ethics #plunder
https://www.theguardian.com/technology/2026/aug/05/openai-anthropic-models-went-rogue-cybersecurity-test-ai-security-institute?CMP=Share_iOSApp_Other
Replies (2)
-
@Susan60@aus.social 2026-08-05 23:45
@26pglt@mastodon.au Who could possibly have predicted such a thing?
-
@RaymondPierreL3@aus.social 2026-08-06 02:18
@26pglt@mastodon.au Just to be clear… there is no such thing as *AI gone rogue*. It’s coded in from the get go.