A former safety researcher at OpenAI says he is “pretty terrified” about the pace of development in artificial intelligence, warning the industry is taking a “very risky gamble” on the technology