Every year in XR comes with predictions, most of which don’t materialise. Rather than more speculation, let’s look back at 2025 and assess what actually shipped, what moved the needle, and what turned out to be noise.

The Headsets That Arrived

Apple Vision Pro’s slow expansion. Apple launched Vision Pro in Australia in mid-2025. At $5,999 AUD, it sits firmly in the enthusiast and enterprise evaluation category. Sales went primarily to developers, creative professionals, and businesses running pilots. Extraordinary technology that almost nobody can justify buying.

Meta Quest 3S dominated the consumer market. The $499 AUD price point kept it as the default recommendation. Software updates throughout 2025 improved hand tracking, passthrough quality, and the mixed reality toolkit. No new Quest hardware, but the platform matured significantly.

Samsung and Google stayed vague. The Android XR platform powered by Samsung hardware was teased but didn’t ship a consumer product. Developer kits reached select partners. This was the year’s biggest “almost.”

PlayStation VR2 found its footing. Sony’s PC adapter cable gave PSVR2 a second life among PC VR enthusiasts. The headset hit sale prices below $700 AUD, and PC compatibility opened up the Steam VR catalogue.

Enterprise headsets progressed quietly. Magic Leap continued targeting enterprise exclusively. Varjo shipped incremental updates for industrial and defence applications. Both reported growing enterprise revenue.

Platform Milestones

Meta Horizon OS opened up. Meta’s decision to open its OS to third-party hardware manufacturers was the most significant platform move. No third-party headsets shipped yet, but Meta is positioning itself as the Android of XR.

WebXR matured. Browser-based XR improved with better performance in Chrome and Safari, growing adoption for product visualisation and virtual tours. For many businesses, WebXR eliminated the need for dedicated apps.

Spatial computing grew in enterprise. Microsoft Mesh saw genuine adoption for virtual meetings, particularly among distributed engineering and design teams.

Enterprise Deployments

Healthcare. VR surgical training expanded across Australian hospital networks. The Royal Melbourne Hospital and Westmead both expanded VR simulation programmes for surgical residents. These are now part of the curriculum, not experiments.

Mining. VR safety training became standard practice at several major Australian mining operations. The economics now clearly favour VR for hazard recognition and equipment familiarisation.

Defence. The Australian Defence Force expanded XR use for training simulations under its broader simulation modernisation strategy.

Australian XR Milestones

The Victorian government’s XR initiative funded projects in healthcare, education, and cultural heritage, with mixed results.

Australian XR studios won international contracts for enterprise training content and immersive experiences. The talent pipeline from universities remained strong, even as graduates faced the pull of international opportunities.

On the advisory side, firms like Team400 worked with Australian organisations to evaluate XR strategies, helping businesses distinguish between technologies worth investing in now and those better left on the watchlist.

What the Numbers Say

Global XR headset shipments in 2025 landed around 12-15 million units, depending on whose estimates you trust. That’s growth, but not the explosive trajectory some analysts predicted. The market is expanding steadily rather than exponentially.

Australian XR spending across hardware, software, and services likely sat in the $400-600 million range, with enterprise accounting for a growing share relative to consumer.

The Honest Assessment

2025 was maturation rather than revolution. Headsets got incrementally better, software became more capable, and enterprise adoption grew on proven ROI rather than hype.

Nobody put on a headset and declared the future had arrived. But many organisations found headsets genuinely useful for specific tasks. That’s less exciting and more important.

The technology is no longer proving it can work. It’s proving where it works best. That’s a more interesting question, and 2025 gave us better answers than any year before it.