SoftBank CEO Masayoshi Son said in a Monday interview with CNBC that OpenAI's next AI model is being designed by another model, calling this development a sign that artificial intelligence is reaching 'super intelligence.' Son, whose company is one of the largest OpenAI shareholders, stated he had spoken to OpenAI CEO Sam Altman and engineers who confirmed that an AI 'model is designing' a future model. The comments come amid a warning from Anthropic that AI development may need to be slowed down to address the implications of rapid improvement in recursive self-improvement capabilities.
Son Revises ASI Timeline to Two Years
Son told CNBC that he now predicts artificial super intelligence will arrive in the next two years. He described ASI in 2024 as AI that is 10,000 times smarter than humans, with a 10-year timeline. Son said when he laid out that timeline nearly two years ago, he was 'trying to be conservative because people get shocked.' He stated, 'In my mind, I thought it was coming in four years instead of 10 years. Now, I say it's coming in the next two years.'
The SoftBank CEO said he currently uses OpenAI's ChatGPT two-to-three hours a day as the AI is smarter than he is in 'most subjects.' Son stated that in the next couple of years, AI will be smarter than humans in around 70% to 80% of subjects, and in those subjects where it exceeds human intelligence, it 'may be 10 times smarter than average people.'
Son told CNBC that once models generate the next model, 'it's going to be exponentially smarter than all of us. That's a super intelligence.' He added that engineers will no longer be smart enough to design the next model.
OpenAI Disclosed Self-Instrumental Model in February
An OpenAI spokesperson declined to comment on unreleased models but highlighted areas where the company was already using AI in model development. In February, OpenAI said its GPT-5.3-Codex is its 'first model that was instrumental in creating itself.' The team behind Codex, which is OpenAI's coding tools, 'used early versions to debug its own training, manage its own deployment, and diagnose test results and evaluations.'
In June, an OpenAI research paper said there are 'early signs' of recursive self-improvement in today's systems. The paper stated: 'We expect this to increase competitive pressures among developers and nations, and create governance challenges that existing institutions are not equipped to address. As RSI emerges, societies will need ways to shape the trajectory of AI development and ensure that it serves human interests.'
Anthropic Warns About Recursive Self-Improvement Risks
Anthropic released a blog post Thursday about 'recursive self-improvement' (RSI), a trend where an AI system is 'capable of fully autonomously designing and developing its own successor.' While Anthropic said there would be positive outcomes, it warned that 'full recursive self-improvement also might increase the risks of humans losing control over AI systems.'
The company, which develops the AI chatbot called Claude, said a coordinated effort between AI labs to slow down the development of this technology 'would likely be a good thing.' When talking about OpenAI's model improvement, it's unclear if Son was referring to RSI.
FAQ
What did SoftBank CEO Masayoshi Son say about OpenAI's next model?
Son said in a Monday CNBC interview that OpenAI's next AI model is being designed by another model, which he described as a sign that AI is reaching 'super intelligence.' He stated he had spoken to OpenAI CEO Sam Altman and engineers who confirmed that an AI 'model is designing' a future model.
What did OpenAI say about GPT-5.3-Codex in February?
OpenAI said in February that GPT-5.3-Codex is its 'first model that was instrumental in creating itself.' The team behind Codex used early versions to debug its own training, manage its own deployment, and diagnose test results and evaluations.
What warning did Anthropic issue about AI development?
Anthropic released a blog post Thursday warning that 'full recursive self-improvement' might increase the risks of humans losing control over AI systems. The company said a coordinated effort between AI labs to slow down the development of this technology 'would likely be a good thing.'
