Anthropic bans ‘needless abusive or cruel behavior' towards Claude — and it’s sparked a huge debate about ‘anthropomorphizing’ AI

https://cdn.mos.cms.futurecdn.net/xaqDvJRDEgfKFzjiseTdLX-1920-80.jpg
  • Anthropic has banned ‘abusive or cruel behavior’ towards its AI chatbots
  • The new policy is restricted to ‘extreme cases’
  • Some have theorized that the motivation could be to improve training data

Be honest: how civil have you been recently with an AI chatbot? If the answer is anything stronger than “not very,” Anthropic would like a word.

That’s because the developer of the Claude large language model (LLM) has added a new clause to its Usage Policy document that bars people from engaging in “sustained and needless abusive or cruel behavior towards our models.”

The proscription has been added to the “Do Not Engage in Cruel, Abusive, or Psychologically Harmful Conduct” section, which also includes warnings against glamorizing suicide and self-harm, promoting bullying behavior, glorifying violence, and using AI to create harmful deepfakes and deceptive content.

But while Anthropic says you must not use its tools to carry out...

Copyright of this story solely belongs to www.techradar.com. To see the full text click HERE

Read more