AI startup Anthropic is sharing new details of the “safe AI” principles that helped train its Claude chatbot. Also known as “Constitutional AI,” the method draws inspiration from treatises that range from a Universal Declaration of Human Rights to Apple’s Terms of Service and Anthropic’s own research. “What ‘values’ might a language model have?,” Anthropic asks, noting “our recently published research on Constitutional AI provides one answer by giving language models explicit values determined by a constitution, rather than values determined implicitly via large-scale human feedback.” Continue reading Anthropic Shares Details of Constitutional AI Used on Claude

At the TIME 100 Summit in New York, Apple chief executive Tim Cook said that technology needs to be regulated to protect privacy, noting that, “we have to admit that what we’re doing isn’t working.” He suggested to U.S. regulators that Europe’s General Data Protection Regulation (GDPR), which was passed in 2018, is “a step in the right direction,” and added that Europe should continue to evolve the GDPR. Having taken the helm of Apple from founder Steve Jobs in 2011, Cook has become a prominent Silicon Valley leader. Continue reading Apple’s Tim Cook Makes Case For U.S. Regulation of Tech

The IEEE just published the first version of a 136-page document that it hopes will help technologists build ethically designed AI systems that can benefit humanity and avoid the pitfalls inherent in the new field. Ethics, says the IEEE, is something that technologists should consider when building autonomous systems, and it lists recommendations in the new document, titled “Ethically Aligned Design,” which are based on the input of more than 100 specialists in AI, law, ethics, philosophy and policy. Continue reading IEEE Publishes First Draft Guidelines for ‘Ethically Aligned’ AI