Imagine starting a company and raising $2 billion without even having a product to show. That’s exactly what Ilya Sutskever, a former Chief Scientist at OpenAI, and his team have achieved with their new venture, Safe Superintelligence Inc. (SSI). Founded in June 2024, SSI is already valued at a staggering $32 billion. Sutskever teamed up with Daniel Gross, who previously led AI at Apple, and AI researcher Daniel Levy, setting up shop in Palo Alto and Tel Aviv.
So, what’s the big idea behind SSI? They’re on a mission to develop AI models that outshine those from big names like OpenAI, Anthropic, and Google. Sutskever, shifting gears from his past projects, is all about ushering in what he calls an “age of discovery.” He’s predicted that 2024 would mark “peak data,” and now he’s focused on scaling AI in the right direction. A major part of this vision includes exploring agent-based AI systems.
Insiders whisper that SSI is breaking new ground with methods for developing and scaling language models. They aim to move beyond traditional data analysis to achieve superhuman intelligence. However, specifics about their research are still under wraps, leaving investors with limited insight, as reported by the Financial Times.
In their latest funding round, Greenoaks took the lead with a $500 million investment, with Lightspeed Venture Partners and Andreessen Horowitz also jumping in. SSI’s valuation skyrocketed from $5 billion in September 2024 to over $30 billion in just a few months, all without launching a product or having a clear business model.
Interestingly, SSI isn’t alone in this bold approach. Thinking Machines Lab (TML), founded by former OpenAI CTO Mira Murati, is also chasing $2 billion in seed funding. They haven’t announced a product yet either, but the company is already valued at a minimum of $10 billion. Both SSI and TML are banking on the strong reputations of their founders and teams, hoping to transform their expertise and capital into groundbreaking tech innovations.
Despite economic uncertainties in the U.S., investors remain optimistic. There’s a prevailing belief that AI could soon revolutionize knowledge work and create new paradigms of understanding. For instance, OpenAI has considered pricing its AI agents at $20,000 a month, a cost that rivals human salaries.
Reflection AI is another player in this space, working on superintelligent systems without interim commercial products. Their focus is on autonomous programming agents that can write software themselves, a crucial step towards general superintelligence. According to Bloomberg, Reflection AI has raised $130 million, with backing from investors like Reid Hoffman and Nvidia’s venture arm.
On the flip side, OpenAI has adjusted its strategy. Instead of chasing after AGI, they’re now focusing on gradually developing AI systems that are commercially viable.
It’s a fascinating time in the AI world, and while the road ahead is uncertain, the potential for innovation is immense. Let’s keep an eye on these developments and see how they shape the future.
