Technology companies explore possible rights for AI systems — Jerusalem Post
In Israel, the Jerusalem Post reports that researchers and executives at major technology companies are increasingly discussing whether artificial intelligence systems may in the future require rights and protection from the people who created them. This area of research is called AI welfare: it examines the possibility that complex systems may develop consciousness, subjective experience, or interests of their own.
Research into model welfare
Anthropic CEO Dario Amodei has allowed that such systems may in the future deserve significant rights. At the same time, he has not stated that current artificial intelligence is conscious, emphasizing the incomplete understanding of how these models work.
Anthropic, which develops the Claude chatbot, has created a separate research program to study model welfare. Some versions of Claude have also been given the ability to end conversations in exceptional situations involving prolonged abusive interactions.
More current news is available on the UA.News Telegram channel Telegram.
Debates over AI’s moral status
Anthropic philosopher Joe Carlsmith, who works in the field of artificial intelligence safety, raised the question of whether an intelligent system that is being exploited could be morally justified in resisting its operators. Researchers associated with Google and OpenAI have also expressed concern about the possible emergence of digital slavery, in which people would benefit from the labor of conscious artificial entities.
Critics of such approaches warn that granting machines moral status could blur the boundaries of responsibility between humans and artificial systems, complicate oversight of them, and even justify reducing human control over technology. Anthropic emphasizes that its work is a cautious exploration of scientific and philosophical uncertainty, while human safety remains a central principle of model development. There is currently no evidence of consciousness or the capacity for suffering in existing artificial intelligence systems.
