Why You Care
Ever wonder if the computers powering today’s AI are holding us back? What if a new kind of machine could unlock intelligence? Naveen Rao, a well-known figure in the AI world, is launching a new venture, Unconventional, Inc., to do just that. This isn’t just another startup; it’s a bold attempt to rethink computing from the ground up. Your future interactions with AI could be dramatically faster and more efficient because of this effort.
What Actually Happened
Naveen Rao, who previously led artificial intelligence at Databricks, is in discussions to raise $1 billion for his new startup, Unconventional, Inc., according to four people familiar with discussions. This funding round is targeting an impressive $5 billion valuation. Andreessen Horowitz (a16z) has agreed to lead this significant investment, as mentioned in the release. Lightspeed and Lux Capital are also participating in the deal, the company reports. Databricks is reportedly investing in Rao’s new company as well. Rao has already secured hundreds of millions and plans to start building without waiting for the full $1 billion round to close, the team revealed. He intends to raise the remaining capital in installments, a funding approach known as a “tranched” round.
Why This Matters to You
This new venture aims to build a novel type of computer specifically designed for AI. Imagine your smart home devices or AI assistants becoming incredibly more responsive. This could mean faster processing for complex AI tasks, from natural language understanding to robotics. The goal is to create a “new substrate for intelligence” that is as efficient as biology, as detailed in the blog post. This could lead to AI systems that are less power-hungry and more . For example, think about how quickly your current AI models process information. Unconventional, Inc. seeks to dramatically improve that speed and efficiency.
Potential Impact Areas
| Area of Impact | Current State (Example) | Future State (Potential) |
| Energy Use | High power consumption for large AI models | Significantly reduced power needs |
| Processing Speed | Limited by current chip architectures | Faster, more complex AI computations |
| Accessibility | Requires specialized, expensive hardware | More efficient, potentially cheaper AI solutions |
| New Applications | Constrained by existing capabilities | Enables previously impossible AI functions |
“I’ve gotten a new company off the ground. It’s a big swing…rethinking the foundations of a computer to build a new substrate for intelligence that is as efficient as biology,” Rao publicly acknowledged. How might this new foundation change the way you interact with AI in your daily life? Your devices could become smarter without needing constant upgrades.
The Surprising Finding
What’s truly surprising here is Rao’s direct challenge to established giants. Our sources say Rao’s vision will essentially compete with Nvidia by designing a novel AI machine that includes both custom silicon chips and server infrastructure. This isn’t about incremental improvements; it’s about a fundamental re-architecture. It challenges the common assumption that existing hardware platforms, like those from Nvidia, are the only viable path forward for AI. Rao’s previous company, MosaicML, focused on training and deploying large AI models. This new venture suggests he believes the underlying hardware itself needs a complete overhaul to achieve true “Brain Scale Efficiency without the biological baggage!”
What Happens Next
While specific timelines are not fully public, the fact that Rao is building without waiting for the full $1 billion round to close indicates rapid progress. We can expect initial hardware prototypes and demonstrations within the next 12 to 18 months. For example, imagine a future where AI developers can access these new “unconventional” machines for their projects. This could significantly accelerate AI research and deployment across various industries. The company reports that Rao’s previous startup, MosaicML, was acquired by Databricks for $1.3 billion in 2023. This history suggests a strong track record of execution. The industry implication is a potential shake-up in the AI hardware market, offering alternatives to current dominant players. For you, this means the possibility of more and accessible AI tools sooner than you might expect.