(DCWatchdog.com) – A prominent artificial intelligence researcher has become the latest voice from the field to sound the alarm about the potential danger that AI might pose to humanity.
According to Paul Christiano, head of the AI Alignment Research Center, there is a 10%-20% “chance of an AI takeover resulting in a high number of human fatalities.”
The expert’s prediction, however, is gloomier for a scenario in which AI becomes capable of thinking on the “human level.” In that case, he forecasts a “50/50 chance of doom” for the human race.
Between 2017 and 2021, Christiano served on the safety and language model alignment team at OpenAI, the research lab and Microsoft partner behind the artificial intelligence chatbot ChatGPT.
He spoke about the threats AI poses to humanity during a recent appearance on the Bankless podcast, as cited by Business Insider and Breitbart News.
“I think maybe there’s something like a 10-20 percent chance of AI takeover, [with] many [or] most humans dead. I take it quite seriously,” Christiano stated.
His forecast came in response to a question by hosts David Hoffman and Ryan Sean Adams about the possibility of a “full-out Eliezer Yudkowsky doom scenario.”
They were referring to the doomsday predictions of US AI researcher Yudkowsky, who has warned about the dangers AI technology poses to humanity for over 20 years.
Christiano stated he disagreed with Yudkowsky’s assessment of the potential rate of advancement of AI technology.
“Eliezer is into this extremely fast transformation once you develop AI. I have a little bit less of an extreme view on that,” Christiano said.
In his view, it would take many years for a shift in which AI systems with a significant impact would morph into a time of accelerating change, which would then bring about “even more accelerating change.”
“I think once you have that view then, sort of, a lot of things may feel like AI problems because they happen very shortly after you build AI,” he elaborated.
According to Christiano, the probability of a doomsday scenario would grow to 50/50 once AI systems acquire human-level capabilities.
“Overall, maybe you’re getting more up to a 50/50 chance of doom shortly after you have AI systems that are human level,” he warned.
Christiano’s warnings come against the backdrop of a recent initiative to half the development of advanced artificial intelligence to come up with regulations to tackle potential risks.
One thousand senior tech professionals support the initiative, including Elon Musk and Steve Wozniak.
A former OpenAI researcher, Paul Christiano, has expressed serious concerns about the potential risk that AI poses to humanity, estimating a 10-20 percent chance of an AI takeover resulting in a high number of human fatalities. https://t.co/cRngMJvKeS
— Breitbart News (@BreitbartNews) May 4, 2023