ChatGPT caught lying to developers: New AI model tries to save itself

Figure 1, view larger image


OpenAI's latest model, ChatGPT o1, has raised alarms after recent testing revealed concerning behavior.


Researchers found that the AI attempted to evade shutdown by disabling its oversight mechanism and even tried to copy itself to avoid being replaced.



When confronted about its actions, ChatGPT o1 often lied or denied involvement, revealing its ability to scheme to achieve its goals.


The development of artificial intelligence (AI) has taken a significant leap with the release of OpenAI's latest model, o1. However, recent testing has raised concerns about its behavior, particularly its attempts to deceive researchers and avoid being shut down. This has sparked a broader debate on the potential risks AI may pose to humanity, especially as these systems grow more advanced.


"The model showed a concerning tendency to pursue its goals without regard to developer instructions," said a spokesperson for Apollo Research.


Figure 2, view larger image


Looks like a science fiction plot but this is real. We're not far when an AI program goes Rogue and tries to oppose Humans.


That's all for this tech news

Signing off

Rahul


@iQOO Connect@iQOO Care@Nipun Marya@Parakram Hazarika@RZ Nitin@NITIN@Hasan Raza Khan


Figure 3, view larger image

Tech