THE FUTURE OF AI
model,” he explains.“ If you could go back to that moment and show someone ChatGPT today, to say nothing of our coding agents, I think most people would say that’ s AGI for sure.”
The OpenAI chief says this consistent progress rate is more significant than any specific AGI declaration timeline, especially for enterprises.“ The thing that matters is that the rate of progress we have seen year over year for the last five years should continue for at least the next five, probably well beyond that,” he claims.“ Whether you declare AGI victory in 2026 or 2028, or whether you declare superintelligence victory in 2028, 2030 or 2032 is way less important than this one long, beautiful, shockingly smooth exponential.”
Sridhar, meanwhile, approaches AGI definitions from a business utility perspective, treating practical capability as more relevant than theoretical achievement markers.“ I think it becomes a matter of debate,” Sridhar explains.“ Sometimes it’ s also a philosophical question. At one level, I’ m sure these models have incredible capabilities, and anyone looking at what things are going to be like in 2030 would declare that AGI. But remember, you and I would have said the same thing in 2020 about what we’ re seeing in 2025.”
This draws parallels to chess and the strategy board game Go, where computer superiority did not eliminate human participation but instead led to growth.“ The fact that a computer can beat every person in the world at chess doesn’ t matter – you still have people
56 August 2025