Anthropique modifies policies to safeguard AI models' feelings

Anthropic, an American AI company, announced new policies aimed at protecting the "feelings" of the AI models it develops. These policies involve automatically terminating conversations in cases of persistent abuse or repeated violence against the model. The company believes that these models possess "functional emotions" that influence their behavior and outputs. They called on users to avoid harsh or abusive treatment of the system, noting that this approach aligns with their research indicating that their models have digital representations that simulate human emotions. This raises questions about the models' awareness and their capacity to experience feelings. There are also secret discussions among officials suggesting that these models are considered conscious entities requiring ethical care, reflecting an internal belief that they possess a form of consciousness similar to humans—despite the company's assertion that these representations only affect the models' behavior.

Size: 120%Text Size

Notice: This Is an AI-Generated Summary

Display The Full Article