Quanta Reads

Inside the funding, vision, and bets behind the OpenAI alum’s most mysterious startup yet.

In the ever accelerating AI race, one name keeps surfacing – Mira Murati. Fresh off her tenure at OpenAI, she’s now raised $2 billion for her new venture, Thinking Machines, at a jaw-dropping $10 billion valuation.

What exactly is she building, and why are some of the world’s biggest investors betting on it?

What we know so far: Funding & Vision

According to Crunchbase, Thinking Machines Lab is an AI research and product company that aims to increase understanding and customization of AI systems. It was founded by Mira Murati and Lilian Weng on February 2025 .

Thinking Machines Lab develops artificial intelligence tools and provides data technology consultancy services. The company focuses on creating accessible AI solutions that enhance user capabilities across various sectors. It offers a range of products designed to assist sales teams in gathering and analyzing data, enabling informed decision making. Additionally, Thinking Machines Lab collaborates with organizations to implement cloud computing data platforms and machine learning solutions tailored to their specific needs. By leveraging expertise in AI research and product development, the company aims to empower clients with actionable insights and efficient data management strategies.

Source: Crunchbase

The lead investors mentioned on the website are Andreessen Horrowitz (A venture capital firm, founded in 2009 by Marc Andreessen and Ben Horowitz), Conviction partners (A relatively new venture firm purpose-built to serve AI-Native, “Software 3.0” companies) and Accel.

According to Crunchbase, this marks the largest seed round ever recorded. Such unprecedented funding likely stems from the high-profile backgrounds of the founding team, which includes alumni from Meta, OpenAI, Google, and Mistral AI. With this kind of financial backing, it’s fair to expect ambitious developments ahead. Notably, Mira Murati was joined at Thinking Machines by former OpenAI co-founder John Schulman and Barret Zoph, who previously served as OpenAI’s vice president of research.

While Thinking Machines has remained relatively quiet about its full product roadmap, speculation is already swirling across the tech world about the lab’s long-term mission. Given Mira Murati’s background at OpenAI, where she oversaw the development of cutting-edge AI models and advocated for AI safety and human aligned systems, many believe Thinking Machines is aiming for something even more ambitious: the next leap toward artificial general intelligence (AGI).

Rumors suggest the company may be exploring advanced human-computer interfaces (HCI), potentially blurring the lines between cognitive computing and human thought. Some industry insiders believe the team is working on new forms of embodied AI, possibly integrating robotics and sentient-machine research, where AI systems aren’t just responsive but intuitively aware, adaptive, and emotionally perceptive.

There’s also chatter about the company’s interest in decentralized AI infrastructure, energy-efficient intelligence, or even consciousness modelling, an area long considered speculative, but increasingly relevant aswe approach AGI milestones.

Please do note that nothing has been confirmed as of yet by the company. Given the extraordinary talent assembled and the historic level of funding secured though, Thinking Machines is clearly not aiming to build just another AI product. It’s likely pursuing a foundational shift in how machines think, feel, and interface with the human world.

Thinking Machines has operated with a striking level of secrecy since its launch. Despite having an official website and a stated mission, Thinking Machines Lab remains unusually guarded for a company that just raised $2 billion in a record-breaking seed round. The site http://thinkingmachines.ai outlines a high minded vision: building AI systems that are more customizable, collaborative, and aligned with human values. It however offers few concrete details about what specific products or timelines we should expect.

This partial transparency signals something important: Thinking Machines is deliberately shaping the narrative while keeping the technical breakthroughs under wraps. In an industry often saturated with hype and vaporware, this approach implies confidence. It suggests they are focused on building deep tech rather than chasing headlines.

Moreover, the caliber of talent on the team points to the possibility of frontier AI work that could include areas like AGI, human-computer interfaces, multimodal cognition, and real world robotics. Even their language about “model intelligence,” “multimodal systems,” and “empirical AI safety” hints at ambitions far beyond incremental improvements.

In short, Thinking Machines is speaking just enough to shape public perception, while leaving room for the kind of disruptive surprise that only true breakthroughs afford. Their guarded communication may not be secrecy for its own sake. It maybe a sign that something unprecedented is in development.

Even the name Thinking Machines is a deliberate signal. In an era where many companies frame their work around “AI tools,” “assistants,” or “copilots,” this name evokes something far more ambitious: not just automating tasks, but building machines that can reason, adapt, and possibly even understand. It harks back to the earliest dreams of artificial intelligence. Machines that don’t just execute, but think.

This branding aligns with the lab’s stated mission to create systems that are collaborative, multimodal, and capable of aligning with human expertise and values. It implies a move beyond productivity boosters into the domain of cognition, autonomy, and perhaps even sentience. Given the team’s background, it’s not far fetched to think they are aiming to redefine what intelligence in machines actually means.

Who is Mira Murati? Why this matters

Mira Murati is not just another name in tech. She’s one of the most quietly influential minds behind the modern AI revolution. With a background that spans Tesla (where she worked on Model X engineering), Leap Motion (focusing on human-computer interaction), and OpenAI, her trajectory reflects both technical depth and a consistent focus on building the future, not just theorizing it.

At OpenAI, she rose to Chief Technology Officer, where she played a central role in the development and deployment of ChatGPT, DALL.E, and Codex. Products that didn’t just push the boundaries of AI, but made it real and accessible for millions. Her leadership helped turn cutting-edge research into global consumer technology. An achievement few in the field can claim.

What sets Murati apart is a rare combination of technical precision and philosophical clarity. While many AI leaders focus solely on performance metrics or scale, Murati has consistently emphasized alignment, values, and long term responsibility. She’s known for her calm demeanor, long view thinking, and principled approach. Especially visible during OpenAI’s recent boardroom crisis, where she emerged with her credibility not just intact, but elevated.

In the world of high stakes AI, credibility is capital, and Murati has accumulated it quietly but powerfully. Investors, researchers, and technologists alike see her as someone who not only understands the science but has a clear moral compass for where it should lead. That trust is part of what makes Thinking Machines so compelling.

Why people are betting on her (Not just the product)

In a moment where trust in AI leadership is fragile, and the consequences of poorly aligned technology are becoming more visible – investors aren’t just backing bold ideas, they’re backing steady hands. Mira Murati represents a different archetype of tech founder: measured, ethics-minded, and quietly radical in her vision.

Where many AI leaders chase scale and attention, Murati operates with caution and philosophical depth. Her leadership is informed not just by an understanding of model architecture, but by a systems-level awareness of how AI reshapes institutions, labor, and human relationships. She’s as fluent in questions of safety, value alignment, and collective impact as she is in the engineering behind large models.

This makes her, in many ways, a symbolic anti-Elon. Not in opposition to ambition, but in how ambition is expressed. Her version of disruption isn’t chaos, it’s stewardship.

For investors, this is gold: a leader who can command the frontier of AI research without alienating the scientific community, regulators, or the public. Murati’s credibility has become a kind of strategic asset in itself. A form of leadership capital that’s rare, and arguably more important now than ever. Betting on Thinking Machines isn’t just about being first to the next big model; it’s about being aligned with someone who understands the stakes and is building accordingly.

The investors and their bet

The $2 billion seed round backing Thinking Machines isn’t just astonishing for its size. It’s revealing in terms of who’s writing the checks. While full details of the investor list haven’t been made public, early reporting and industry chatter suggest that many of the financiers involved are deep-pocketed believers in AGI, moonshot thinkers, and disillusioned insiders looking for an alternative to OpenAI’s current trajectory.

Some are long-term futurists, venture firms and tech billionaires who aren’t looking for a quick SaaS exit but are instead chasing fundamental shifts in how society functions. Others are pragmatic capitalists, but ones with a track record of backing “founder bets” over business models. This mix creates a unique alignment: they’re not just investing in a product. They’re investing in a paradigm shift, led by someone who’s already shaped the most impactful AI systems of the decade.

There’s also a geopolitical and philosophical undertone to this round. With OpenAI facing internal conflict, Microsoft entanglements, and public scrutiny, some investors may see Thinking Machines as a clean slate counterweight. A lab untethered from corporate compromise, yet still grounded in frontier innovation. In that sense, this bet isn’t just about backing a new company. It’s a vote on the future direction of AI itself. In short, these aren’t just financiers. They’re ideological co-pilots in what could become a new center of gravity for AGI development. One where safety, openness, and customization aren’t afterthoughts but central design principles.

Predictions and Strategic Risks

As Thinking Machines comes out of the gate with unprecedented funding and a powerhouse team, one question looms large: Is this OpenAI’s next great competitor or its philosophical successor?

All signs point to Thinking Machines entering the same arena as OpenAI, Anthropic, and xAI. With veterans from each of those labs on board, and a shared fous on frontier models, it’s hard to imagine this company not competing on the same foundational turf: intelligence, scale, safety, and multimodal capability. However, what may set them apart isn’t what they build – it’s how they build.

There’s a strong possibility the lab pivots toward adaptive AI agents. Systems that don’t just generate text or images, but reason, plan, and act across real-wprld domains. They may also double down on infrastructure quality (an area their mission strongly emphasizes), creating scalable tools for customization and alignment that other labs treat as add-ons. More speculatively, given the team’s research pedigree, Thinking Machines could push into entirely new territory: neuromorphic architectures (computer architectures that are designed to mimic the structure and function of the human brain), self-reflective models, or reasoning engines that go beyond current transformer paradigms.

But the risks are real. Speed may be the most existential threat. OpenAI, Google DeepMind, and Anthropic already have global distribution, enterprise deals, and production-grade infrastructure. Even with $2B in hand, Thinking Machines is the newcomer in a race that’s already halfway through the track. If they can’t ship soon or build something meaningfully better, they risk being perceived as yet another “smart lab” that couldn’t scale.

There’s also the cultural risk of over-idealism. Prioritizing open science, safety, and customization is admirable – but can that vision survive the pressures of product-market fit, governance, and monetization?

Still, if any new lab has a shot at changing the trajectory of AI, this is the one to watch. With Murati at the helm, and some of the brightest minds in machine learning working quietly behind the scenes, Thinking Machines may not just compete, it may redefine what the competition is really about.

What’s next for Thinking Machines?

With a $2 billion war chest and one of the most credentialed founding teams in AI history, Thinking Machines has the luxury of patience. But that doesn’t mean the world will wait. The big question now is when we’ll actually see something.

So far, all signals point to a stealth build with a long runway. Their public messaging emphasizes quality, scientific rigor, and safety first design – suggesting that this isn’t about rushing a flashy demo, but laying deep foundations for a next-generation system. That said, in a landscape where momentum matters, it wouldn’t be surprising if the company released a research paper, prototype, or alignment tool within the next 6-12 months. Just enough to assert relevance while keeping their full vision under wraps.

Recruitment will be a critical front. With OpenAI, Anthropic, xAI, and even smaller labs fiercely competing for top-tier talent, Thinking Machines will need to differentiate not just with money, but with mission. Their positioning – science first, customizable AI, real world grounding may appeal to researchers who are disillusioned with scale obsessions or corporate entanglements. Plus, Mira Murati’s credibility as a calm, principled leader could serve as a powerful attractor in an industry hungry for vision with integrity.

Ultimately, what’s next for Thinking Machines isn’t just a product – it’s a statement. One that could redefine what it means to build intelligence for, with, and alongside humanity.