Current Average Estimate for AGI

Many experts in the field of artificial intelligence (AI) have provided their estimates for when they believe artificial general intelligence (AGI) will be achieved. By averaging these estimates, we have created a countdown to AGI. Please note that when date ranges are given as estimates, we use the lower end for our countdown, resulting in a slightly more aggressive average estimate.

Since the definitions of AGI can vary, we have left this countdown open to each contributor’s own interpretation of the term’s precise meaning.

The current estimate, based on three contributions, is October 1st, 2025.

Using https://onlinetimetools.com/calculate-average-calendar-date

This countdown should not be considered a rigorous scientific approach to estimating the arrival of AGI or a formal approach to the topic. AGI itself is not well defined and may not even have a precise definition. However, it is an interesting topic worth exploring.

Contributors

Below is the list of current contributors (we have more invitations pending via Twitter DMs) to the estimate, along with any thoughts they have shared:

Jacob Valdez – December 31st, 2023

Elon Musk? (invitation for an estimate has been sent)

Robert Scoble? (invitation for an estimate has been sent)

Geoffrey Hinton – April 1st, 2028 (source: twitter.com/JMannhart)

Steve Digital – January 1st, 2025

“2022 and the previous years have witnessed significant progress in AI. In my opinion, 2023 will surpass those advancements, and with further improvements in emergent behaviors and abilities, I expect AGI to be a reality by early 2025. This is an optimistic timeline. However, I am cautious about the implications of AGI for humanity, although I hope it will lead to significant improvements in people’s lives worldwide.”

Other Contributions

Despite some feedback claiming that this article is either ridiculous or lacks a clear definition to estimate a date or range, we still wanted to explore the topic. Here are a few examples: