One of the most discussed topics in recent months is the malicious use of artificial intelligence chatbots. By following a few guidelines, most chatbots, including ChatGPT, can produce pretty brutal results when ignoring security rules. This problem can be solved with a new artificial intelligence developed by a group leaving the OpenAI team, the creators of ChatGPT.

Claude, a chatbot created by Anthropic, allegedly adheres to a number of rules, including the Universal Declaration of Human Rights. The firm says chatbots have a “conscience”.

Moral artificial intelligence

According to the news in Wired, Claude revolves around a structure that the developers call the constitution, which includes many important rules of civilization, such as the Universal Declaration of Human Rights. “It basically reinforces unconstitutional behavior and deters problematic behavior,” said Jared Kaplan, founder of Anthropic.

While the leading names of technology, including Elon Musk, say that the rapid growth of artificial intelligence may cause problems for people in the future, Anthropic’s solution to this problem seems to be ethical artificial intelligence.

Said to be more inclined to give answers that support freedom of thought, conscience, opinion, expression, assembly and religion, equality and fraternity, Claude is brought up within a strict list of rules. “This seems like a great idea for Anthropic, which leads to good experimental results,” said University of Washington researcher Yejin Choi. According to said Choi, we should create such constitutions and norms for the future of artificial intelligence.