My AI Timelines Have Sped Up

A Software Engineer on the Robotics at Google team revises his estimates as to when artificial general intelligence could happen.

In 2015, I made the following forecasts about when AGI could happen.

  • 10% chance by 2045
  • 50% chance by 2050
  • 90% chance by 2070

Now that it’s 2020, I’m updating my forecast to:

  • 10% chance by 2035
  • 50% chance by 2045
  • 90% chance by 2070

The AGI debate is always a bit of a mess, because people have wildly divergent beliefs over what matters. One useful exercise is to assume AGI is possible in the short term, determine what could be true in that hypothetical future, then evaluate whether it sounds reasonable.

Read More at Sorta Insightful

Read the rest at Sorta Insightful