July 3, 2026

Odyssey is now a US$1.45 billion AI company built by a Kiwi founder

Detailed image of a server rack with glowing lights in a modern data center.

A cheque with strings attached

Auckland-educated founder Jeff Hawke has closed a US$310 million (NZ$529m) Series B round for his AI company Odyssey, valuing the business at US$1.45 billion (NZ$2.5 billion). The round was led by US firm Natural Capital, with Amazon, AMD Ventures, the CIA’s In-Q-Tel, Google Ventures and New Zealand’s Icehouse Ventures all in.

The cast list alone tells you this is not charity. Amazon Web Services becomes Odyssey’s preferred cloud provider, and Odyssey is co-optimising its models on Amazon’s own Trainium chips through Annapurna Labs. That is a strategic infrastructure relationship, not a passive investment. As Amazon VP Ron Diamant put it, “world models represent one of the most demanding workloads in AI.” Amazon does not say that about bets it considers marginal.

Not another chatbot

The reason the big names showed up is that Odyssey is chasing a different category. Where OpenAI, Anthropic and Google built large language models that process text, Odyssey is building a “world model” – AI that learns to simulate and predict the physical world.

Icehouse Ventures partner Mason Bleakely draws the line cleanly: “OpenAI built models that help computers understand language. Odyssey is building models that help computers understand the physical world. If they get it right, that becomes core infrastructure for robotics, autonomy, gaming and any industry that needs to simulate real environments.”

Hawke frames it as “taking AI out of the classroom and into the world”. He is not an unknown. After graduating from the University of Auckland in 2009 and completing a doctorate in applied AI at Oxford, he was a founding research engineer at Wayve, the autonomous driving company now valued at US$15 billion. He co-founded Odyssey with Oliver Cameron in 2023.

The money makes sense against the backdrop. Goldman Sachs estimates US$7.6 trillion will be spent globally on AI infrastructure over the next five years, and tech giants are now burning the inflation-adjusted equivalent of the entire Apollo programme every 6 to 10 months. A world model that works is a land grab worth entering early.

Where the value actually lands

Here is the part that should sting. Odyssey is London-based. The New Zealand content is Hawke’s passport, his Auckland degree, and Icehouse Ventures’ slice of the cap table. That is genuinely something, but it is not a New Zealand-headquartered company creating New Zealand jobs, paying New Zealand tax and keeping intellectual property onshore.

We have watched this movie before. New Zealand educates world-class founders and then exports them, along with the compounding economic value they create. The jobs, the IP, the tax base and the supply chain for Odyssey sit in London and on AWS servers, not in Auckland. Celebrating the raise as a Kiwi win is fair. Mistaking it for economic capture is not.

Is the policy moving fast enough

The government at least knows the gap exists. Its July 2025 AI Strategy noted that 68% of NZ SMEs have no plans to evaluate or invest in AI, against just 38% of Australian SMEs, while estimating generative AI alone could add NZ$76 billion to the economy by 2038.

The research settings are shifting too. In April 2026 an 88% funding increase was proposed for AI, quantum, aerospace and synthetic biology, which AI Forum NZ chair Associate Professor Mahsa McCauley said “directly addresses structural imbalance where advanced technology funding sits at around half the average for small advanced economies.” KiwiNet CEO Dr James Hutchinson welcomed a proposed ARPA-style Ignition Fund that “could unlock bold, high-risk research New Zealand has historically been too cautious to pursue.”

A dose of realism is warranted at the adoption end. Victoria University AI researcher Andrew Lensen warned in May 2026 that “most studies into AI’s economic impact show mixed results or negative return on investment”, with too many firms “simply enabling Copilot and declaring themselves AI-enabled.”

That caution cuts both ways. The Odyssey raise is the rare case where a Kiwi is building foundational deep tech that Amazon and Google want a stake in, not just consuming an American model on subscription. The open question for business is whether the recent policy moves are fast and deep enough that the next Jeff Hawke builds here rather than merely graduating here. On current form, we produce the talent and someone else banks the platform.

Sources

Subscribe for weekly news

Subscribe For Weekly News

* indicates required