When is it permissible for artificial intelligence to lie? A trust-based approach

Open in new window