Musk Estimates Just 10-20% Chance of AI Doom, But Experts Warn Risks Remain High
-
Elon Musk estimates a 10-20% chance that AI could end humanity, but says AI advances should still be explored as positive outcomes may outweigh negative ones.
-
An AI safety researcher believes the risk of AI ending humanity - referred to as "p(doom)" - is actually 99.999999%, saying AI should not be built at all to prevent this.
-
Both researchers and executives typically estimate the p(doom) risk between 5-50%, much higher than Musk's estimate.
-
Musk has raised concerns about AI advances from companies like OpenAI, citing a lack of guardrails and public accessibility.
-
Musk also warns that AI and robotics are on the cusp of a huge revolution, stating "we live in the most interesting of times."