WWDC26 starts on June 8, 2026 and we are about to celebrate three years of visionOS. At Studio Meije, we have a lot of expectations for visionOS 27, from deeper AI integration with the new Siri to significant updates to key frameworks like RealityKit.
A lot has changed over the past three years, and especially over the past six months:
- The AI revolution is moving faster than many of us anticipated and is already changing the way we think about software engineering, product management, and even how companies are built and managed.
- Major political events are having a clear impact on supply chains, and the overall economy is not in great shape right now.
- Glasses as a form factor seem to be capturing consumer attention much faster than expected. We feel this is shifting priorities away from rapid iteration on pass-through headsets like Vision Pro for many companies in the market, including Meta, Google, and probably Apple too. That does not mean the end of the headset form factor though.
- Tim Cook is stepping down as Apple CEO on September 1, 2026. A leadership transition will inevitably have a huge impact on Apple, even if the results will take time to become visible. It will also be interesting to see how John Ternus approaches spatial computing at Apple.
Spatial computing is not a sprint but a marathon, yet navigating the ecosystem without getting impatient can still be challenging. WWDC26 will definitely shape how we approach the next 12 months. We wanted to share a few signals we have been watching closely and what we think could become part of visionOS 27.
Improved USD support
OpenUSD v26.03 was released earlier this spring, bringing support for Gaussian Splats along with broader rendering and performance improvements. More interestingly, a protected post for OpenUSD v26.05 appeared on the Alliance for OpenUSD website a few days ago, while the v26.05 release itself was officially tagged in April. It is a small but interesting sign that the OpenUSD ecosystem is evolving extremely quickly right now.
Apple's relationship with OpenUSD goes far beyond simple file format support. Apple was one of the founding members of the Alliance for OpenUSD alongside Pixar, Adobe, Autodesk, and NVIDIA, and has repeatedly described OpenUSD as a core technology for visionOS and Reality Composer Pro.
Better OpenUSD support is not just about importing assets more reliably. It is about making the entire spatial content pipeline more coherent, from DCC tools to Reality Composer Pro, Xcode, RealityKit, Quick Look, and eventually the spatial web itself. It can seem trivial, but it is still a major friction point on many of our current projects.
Remote Rendering improvements
RemoteImmersiveSpace in visionOS 26 and Foveated Streaming with CloudXR support in visionOS 26.4 clearly show the direction Apple is taking with spatial computing. Apple is investing more and more in technologies that reduce the amount of compute happening directly on the device itself, whether through remote rendering, streaming, or hybrid rendering pipelines.
With CloudXR and foveated streaming, visionOS can now offload extremely demanding workloads to remote GPUs while maintaining low latency and high visual fidelity on the headset. This feels like a very important long-term signal for Apple Glasses. Lightweight glasses will never carry the same thermal and compute envelope as Vision Pro, so building robust cloud rendering infrastructure now makes a lot of sense.
We expect visionOS 27 to push even further in this direction, especially around hybrid rendering between RealityKit and streamed content, lower latency pipelines, and enterprise workflows. In many ways, building for Vision Pro today already feels like the best way to prepare for Apple Glasses tomorrow.
RealityKit update
RealityKit also feels due for a much bigger update at WWDC26. Over the past three years Apple has steadily improved the framework with better animation systems, lower-level rendering APIs, portals, object manipulation, and deeper SwiftUI integration, but the ecosystem is now expecting a more significant leap forward.
The same applies to Reality Composer Pro. It has become a core part of the visionOS workflow when building immersive experiences, but it still feels more like a scene preparation tool than a complete solution comparable to Unity or Unreal Engine.
WWDC26 feels like the right moment for Apple to clearly define what Reality Composer Pro is supposed to become in the long term.
We recently published a blog post about asynchronous world anchors, a specific feature we would love to see in visionOS 27. This would dramatically reduce friction for business customers in critical scenarios.
New Siri & Visual Intelligence
What has been lacking since launch is not just AI features inside visionOS, but a version of Siri and Visual Intelligence that truly understands spatial context: what you are looking at, what is open around you, what object is in front of you, and what you are trying to do across apps, 3D content, and the real world.
With all the recent rumors around the new Siri, we hope visionOS 27 will be the moment where Apple starts turning Vision Pro into the best possible device for contextual AI. For developers, the big question is how much of this intelligence Apple will expose. We do not expect unrestricted camera access for every app, but more public APIs, App Intents, semantic understanding, or Enterprise APIs around Visual Intelligence would already be a major step forward.
Being able to look at a machine, a product, a 3D model, or a training setup and ask Siri to explain, isolate, compare, or trigger actions would be a huge unlock for spatial computing, and exactly the kind of bridge still missing between the real world and the software layer today.
Less friction for our customers
We love seeing the smiles on people's faces whenever they try Vision Pro for the first, or second, or third time. However, for them to reach that moment of enjoyment, we also realize how annoyed and impatient our business customers can get as they have to go through multiple personalization steps. First, they have to provide their eye prescription, watch a tutorial, and then go through multiple eye tracking calibration steps.
This is unfortunately not scalable for many business-critical scenarios where, despite that friction, visionOS still provides a lot of value.
We appreciate the ability to attach a Vision Pro setup to an Apple Account or Apple Business Account, however many business users do not necessarily have the time or desire to perform that extra setup step when time is of the essence.
Some companies use an Accessibility configuration that allows users to control the system cursor using head rotation instead of eye tracking, and we believe this fallback could be promoted more aggressively for quick demos or business presentations. Ideally, we would also appreciate a faster and simpler eye calibration setup, with a single calibration pass instead of multiple ones.
Improved Godot Support
Another interesting signal for the ecosystem is the continued improvement of visionOS support in Godot. Godot already supports exporting windowed apps for Apple Vision Pro, and godotengine/godot#109975 now proposes a dedicated visionOS XR module for building immersive experiences.
One of the stated goals from Apple's visionOS engineering team was not only to support traditional windowed apps on Vision Pro, but also fully immersive experiences through a dedicated visionOS VR plugin.
This is important because it shows Apple investing beyond its own first-party frameworks and tooling. Supporting engines like Godot lowers the barrier for indie developers and experimental spatial apps, and helps broaden the visionOS ecosystem beyond RealityKit-only workflows.
Final words
WWDC26 feels like a potentially defining moment for visionOS. We are expecting visionOS 27 to bring meaningful progress across AI integration, rendering, developer tooling, and enterprise workflows, while also helping reduce some of the friction that still exists today. More than anything, we hope this year's WWDC shows Apple pushing spatial computing forward with confidence and making it clear that Vision Pro and visionOS remain a major long-term priority for the company.
