Australia's Megaport is raising $594 million in fresh capital after securing four new AI infrastructure contracts, the network-as-a-service provider disclosed. The raise reframes Megaport, long known as a connectivity layer between enterprises and the major clouds, as a direct beneficiary of the AI capex cycle. It is one of the largest equity moves by an ASX-listed networking firm tied explicitly to AI demand.
The four contracts are the proximate trigger for the raise, with Megaport pointing to customer-driven capacity needs as the use of capital. The company has not broken out the individual deal values, but the $594 million figure implies a build-out materially larger than its prior expansion cadence. Megaport's core product — software-defined interconnection across more than 900 data centers globally — sits in the path of every GPU cluster that needs to talk to a cloud region, a storage tier, or an end customer.
The structural story is straightforward. AI training and inference workloads have pushed network bandwidth from a procurement afterthought to a binding constraint. Hyperscalers are spending more than $400 billion combined this year on AI infrastructure, and the connective tissue between their regions, colocation partners, and enterprise customers has to scale with them. Megaport sells exactly that connective tissue.
Key facts
- 01Megaport is raising $594 million in fresh capital to fund AI infrastructure buildout.
- 02The Australian network-as-a-service provider has secured four new AI infrastructure contracts.
- 03The raise marks one of the largest equity moves by an Australia-listed networking firm tied to AI demand.
The pivot is also a competitive response. Equinix, Digital Realty, and the hyperscalers' own private fabrics have all moved aggressively to court AI customers with dedicated low-latency links and GPU-adjacent interconnect products. Megaport's pitch is vendor neutrality — a single port that reaches AWS, Azure, Google Cloud, Oracle, and an expanding list of GPU clouds without locking the customer to any one provider. That neutrality matters more when an AI startup is multi-sourcing inference capacity to avoid bottlenecks at any single cloud.
Funding the buildout through equity rather than debt is notable in the current rate environment, where AI infrastructure peers including CoreWeave have leaned heavily on asset-backed financing. The $594 million raise will dilute existing shareholders but gives Megaport a cleaner balance sheet heading into a capex-heavy phase. It also signals confidence that AI-driven revenue will grow into the new capacity rather than stranding it.
Megaport joins a list of infrastructure providers repositioning around AI workloads after the initial wave of capex flowed almost entirely to chips and power. Cooling, networking, and last-mile interconnect are now visibly catching up, with public-market investors willing to underwrite the buildout. The Australian listing makes Megaport one of the few Asia-Pacific-headquartered names trading on the AI infrastructure thesis without being a pure-play data-center landlord.
Execution risk sits on two fronts. The four contracts are concentrated bets — if any one customer renegotiates or slows its AI rollout, the revenue ramp tied to this raise gets pushed out. And the interconnect market is increasingly contested by the hyperscalers themselves, who would prefer customers stay inside a single cloud's private network rather than route through a neutral intermediary. Megaport has to keep proving that neutrality is worth a separate procurement line.
The raise is a data point on how AI infrastructure spending is widening beyond GPUs and into the surrounding stack. Network providers that can credibly attach themselves to named AI workloads are now able to fund growth at scale from public markets, a shift that was not obvious 12 months ago. For Megaport specifically, the question is whether four contracts and $594 million in new capital are enough to lock in a structural seat at the AI interconnect table before the hyperscalers close the gap themselves.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
