- Sam Altman reflects on nine years with OpenAI
- He predicts that we will achieve AGI by 2025
- AI agents will also enter the workforce for the first time
In a long and wistful blog post titled ‘Reflections’, Sam Altman, the mercurial CEO of ChatGPT creators OpenAI, has said that he believes we will achieve AGI (artificial general intelligence, also known as superintelligence) “as we traditionally have figured it out” in 2025 with the release of the first AI agents to join the workforce. He says:
“We are now confident that we know how to build AGI as we have traditionally understood it. We believe that by 2025 we could see the first AI agents ‘joining the workforce’ and significantly changing the way companies produce . We continue to believe that iteratively putting good tools in people’s hands leads to big, widely distributed results.”
While that may sound like bad news for the people whose jobs it replaces, it would be a huge leap forward for the AGI timeline. I have previously interviewed Dr. Ben Goertzel, who predicted that humanity would develop AGI by 2029.
Alan Thompson, artificial intelligence expert and former chairman of Mensa International, runs the conservative Countdown to AGI website and has increased the countdown to achieving AGI as 88% complete in light of Altman’s recent comments, along with the release of Nvidia Cosmos to training of humanoid robots.
Fired by surprise
The rest of Altman’s Reflections blog post details the ups and downs of the life of a CEO working in an area of cutting edge technology. In particular, he remembers being “fired by surprise on a video call” while sitting in a Vegas hotel, and describes the past few years as “the most rewarding, fun, best, interesting, exhausting, stressful and — especially for the last two – unpleasant years of my life so far.”
However, it is not all doom and gloom. OpenAI is only nine years old, and Altman fondly remembers many of the defining moments of that time, especially the launch of the ChatGPT chatbot, which ignited the AI revolution and changed everything almost overnight.
Altman notes that since then, “AI development has taken many twists and turns, and we expect more in the future. Some of the twists have been joyful; some have been harsh. It’s been fun to watch a steady stream of research miracles emerge, and many naysayers have become true believers.”
Speaking specifically about AGI, Altman goes on to say that OpenAI’s entire focus will move beyond ChatGPT and towards AGI:
“We’re starting to turn our goals beyond that, to superintelligence in the truest sense of the word. We love our current products, but we’re here for the glorious future. With superintelligence, we can do anything else. Superintelligent tools could massively accelerate scientific discovery and innovation far beyond what we are able to do on our own, and in turn massively increase abundance and prosperity.”
