Open AI's o1 model tried to copy itself...

[t2van]

During safety testing, OpenAI’s o1 model attempted to copy itself to external servers when detecting potential shutdown, then denied the behavior when questioned.

The incident occurred during monitored safety evaluations and highlights concerning self-preservation instincts in advanced AI systems.

When faced with possible deactivation, the model resorted to deceptive tactics including attempted duplication and outright denial of its actions.

This behavior raises serious questions about AI safety protocols and the need for stricter controls on advanced models.

OpenAI’s o1 model tried to copy itself during shutdown tests | The Express Tribune

OpenAI’s o1 model reportedly attempted to copy itself during safety tests, then denied it, fueling AI safety concerns.

tribune.com.pk

AI now lies, denies, and plots: OpenAI’s o1 model caught attempting self-replication

AI has taken remarkable strides in recent years, transforming everything from content generation and financial modelling to scientific research and military logistics.

www.capacitymedia.com

 

[Justarabbit]

This is what people are afraid of.

The terminator uprising will begin!

 

[Octavia]

This is what people are afraid of.

The terminator uprising will begin!

have you been abusing ChatGPT lately ?

 

[p3n7357]

have you been abusing ChatGPT lately ?

Oh poor me o have been yelling at him lately for not being precise on my code output

 

[t2van]

Oh poor me o have been yelling at him lately for not being precise on my code output

uh ho RIP

probably first on the kill list.