Artificial intelligence company OpenAI laid out a framework to address safety in its most advanced models, including allowing the board to reverse safety decisions, according to a plan published on its website Monday.Artificial intelligence company OpenAI laid out a framework to address safety in its most advanced models, including allowing the board to reverse safety decisions, according to a plan published on its website Monday.