Insecure robots could be better at their jobs

• Autores: Matt Reynolds
• Localización: New scientist, ISSN 0262-4079, Nº. 3129, 2017, pág. 11
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Confidence in their abilities is usually a good thing--as long as they know when it's time to ask for help. As they build ever smarter software, they may want to apply the same mindset to machines. This is the idea behind an experiment that explores a robot's sense of its own usefulness, something that could help guide how future artificial intelligences are made. Overconfident AIs can cause all kinds of problems, says Dylan Hadfield-Menell at the University of California, Berkeley.