A former artificial intelligence insider turned activist has warned of the dangers of rapidly evolving technology, warning that without regulation society is on the verge of crossing a “red line”. Michael Trazzi talks with MS NOW’s Ali Velshi about why he thinks an AI arms race could lead to “things that humans can’t handle.”
“I have been concerned about the risks of AI for about 10 years,” Trazzi told Velshi, explaining that significant progress has been made in the field in that time. “We’ve seen technology advance to the point where we can solve new problems in mathematics and outperform most computer scientists at software engineering,” he said Saturday.
While Truzzi acknowledged that AI has “immense benefits,” he stressed that “with every benefit there is a trade-off,” adding: “While AI can automate our jobs, it can also put our lives at risk.”
Trazzi said a major concern is the race to develop human-level AI, a level of artificial intelligence that replicates or exceeds humans’ ability to learn, reason, and adapt. “As we make these models more powerful, we also make them more dangerous,” he said.
“If AI becomes smarter than any human being on the planet, smarter than any expert in any field, that means we will be able to do things like solve cancer and perhaps develop new biological weapons,” Trazzi said.
“Think about the new coronavirus, but it’s 10 times more deadly and 100 times more contagious.”
“I think this may be a problem that humans cannot deal with,” he added.

