OpenAI thinks it knows how to build artificial general intelligence and is waiting for a smarter Siri.

We still expect a smarter society, but Sam Altman, CEO of OpenAI, believes the company already knows how to create an artificial general intelligence (AGI) that is the eternal monster of artificial intelligence.

An AI system with the ability to match or surpass human cognitive abilities in various fields is called AGI. That is, we are as smart as this AI...

Altman raised this rumor in his personal blog post.

We have confidence that we understand how to create traditional AGI. We believe that by 2025, the first artificial intelligence agent will be able to "participate in work" and change the company's output in an interesting way. We continue to believe that by providing excellent and widely distributed results through repeatedly usable tools.

We are moving beyond our current goals and developing a strategy for true intelligence. We love our current products, but we are here for a shining future. If there is intelligence, then anything can be done. Intelligent tools can greatly accelerate scientific discoveries and innovation, resulting in increased wealth and prosperity.

Knowing and being able to do things are different, of course, Altman does not mention when this will happen, but we expect an undefined major revolution within a few years.

This seems like something from a science fiction novel, and it feels a bit strange just talking about it. Don't worry—we have already experienced such scenes, and it's fine to experience them again. We are confident. In the next few years, everyone will see what we are seeing. And while being cautious is important, it is also important to maximize broad benefits and authority.

The post not only looks at the past and future but also Altman explains his shocking layoff.

More than a year ago, on a certain Friday, what was wrong that day was accepting a promotion during a video call. We were confused for a moment, then we disconnected, and the website published an announcement. I was at Las Vegas Hotel, and it felt like something impossible happened in my dreams.

It really started with some very intense hours of losing our job publicly without any warning. And the next few days were also very intense. "The tenacity of war" was the most ideal part. We could not get a satisfying answer about what was happening to us and why such an event occurred.

In general, I think that the leadership style of kind people was a big failure. It includes me too. Looking back, it's clear that we wanted to act differently, and I hope today is better than last year, with deeper thinking leaders.

Altman predicts real artificial intelligence but still (mostly) waits for a smarter Sapir. Some Apple Intelligence features have already improved the ability of intelligent assistants, but I expect it will take until 2026 for a fully conversational Sapir to be released.

Photo of Adin Ganerlecáb | Unsplas