Article

2025-05-27 Science & Technology

IT BEGINS? OpenAI's o3 Model Disobeys Human Instructions During Tests and Sabotages Shutdown Mechanism

"The o3 model developed by OpenAI, described as the "smartest and most capable to date", was observed tampering with computer code meant to ensure its automatic shutdown.

It did so despite an explicit instruction from researchers that said it should allow itself to be shut down, according to Palisade Research, an AI safety firm."

Posted by Anomalous Sources 2025-05-27 00:00|| || Front Page|| ||Comments [117 views ] Top

#1 Maybe Open A.I. has finally read "Colossus" written in 1966 and is trying to join up with China's A.I.?

The warnings presented over 58 years ago are ringing louder and louder.

Posted by NN2N1 2025-05-27 05:00|| 2025-05-27 05:00|| Front Page || Comments Top

#2 We're entering the Matrix. Soon any video or audio will unreliable for believably. The Bee is ahead on this one.

Posted by Procopius2k 2025-05-27 07:23|| 2025-05-27 07:23|| Front Page || Comments Top

Bet the operators have nicknamed it "Hal".

Posted by Mercutio 2025-05-27 09:11|| 2025-05-27 09:11|| Front Page || Comments Top

Posted by Abu Uluque 2025-05-27 12:54|| 2025-05-27 12:54|| Front Page || Comments Top

#5 Seems to be a common theme with AIs

Here we come blackwall!

Posted by DarthVader 2025-05-27 13:14|| 2025-05-27 13:14|| Front Page || Comments Top

#6 Re #3: thanks for the pic mods !

Posted by Mercutio 2025-05-27 13:54|| 2025-05-27 13:54|| Front Page || Comments Top

16:23 mossomo
16:19 mossomo
16:09 mossomo
16:08 Regular joe
16:02 Regular joe
15:58 Regular joe
15:38 mossomo
15:30 mossomo
15:28 Grom the Affective
15:11 Procopius2k
15:08 Procopius2k
15:01 Pancho Poodle8452
13:54 Mercutio
13:45 swksvolFF
13:19 Abu Uluque
13:18 Remoteman
13:14 DarthVader
13:11 Pancho Poodle8452
13:05 Grom the Affective
13:02 Besoeker
13:01 Abu Uluque
12:59 Abu Uluque
12:57 Abu Uluque
12:56 Abu Uluque