- Updated: March 20, 2026
- 7 min read
AI-Generated Frames Set to Transform Future PlayStation Games

Sony’s next‑gen PlayStation consoles will use AI‑generated frames to deliver smoother gameplay, a technology that creates “imagined” frames between the ones the GPU actually renders.
The move positions PlayStation alongside, and potentially ahead of, PC‑centric solutions like Nvidia DLSS and AMD FSR3, promising a new era of AI‑enhanced gaming on both the upcoming PS5 Pro and the future PS6.
Why AI‑Generated Frames Matter for Gamers
Imagine a game that runs at 60 fps, but feels like 120 fps because the console intelligently fills in the missing moments. That’s the promise of AI frame generation: a machine‑learning model predicts what the next visual slice should look like, then renders it in real time. For tech‑savvy gamers and developers, this means higher perceived fidelity without the hardware cost of a faster GPU.
In a recent interview with Digital Foundry, PlayStation’s lead architect Mark Cerny confirmed that “ML‑based frame generation tech is coming to PlayStation platforms,” hinting at a rollout that could start with the PS5 Pro and later expand to the long‑awaited PS6.
Sony’s Console Roadmap: From PS5 Pro to PS6
Since the launch of the PS5 in 2020, Sony has iterated with the PS5 Pro, a mid‑cycle upgrade that adds a more powerful GPU, faster SSD, and an enhanced AI upscaling engine called PlayStation Spectral Super Resolution (PSSR). PSSR already leverages neural networks to sharpen lower‑resolution images, but it does not create new frames.
The upcoming PS6, teased in Sony’s 2025 roadmap, is expected to feature dedicated “Radiance Cores” for real‑time ray tracing and a larger memory pool. Cerny has hinted that the PS6 may not arrive before 2027, leaving a window for the PS5 Pro to adopt AI frame generation first.
Both consoles share a common hardware foundation—AMD Zen 3+ CPUs and RDNA 3 GPUs—making the integration of a new AI pipeline feasible without a complete redesign.
AI‑Generated Frames vs. Traditional Upscaling
To understand the breakthrough, it helps to compare three approaches:
- Traditional Upscaling (e.g., PSSR, DLSS 2.0): Takes a lower‑resolution image and enlarges it, adding detail via a trained model.
- Interpolation (AMD FSR3 Frame Generation): Calculates intermediate frames by blending adjacent frames—no machine learning involved.
- AI Frame Generation (Sony’s upcoming tech): Uses a deep neural network to “imagine” the content of a missing frame, often producing smoother motion with fewer artifacts.
Because AI frame generation predicts motion vectors and scene depth, it can handle rapid camera pans and complex particle effects better than simple interpolation. However, the technique can introduce a small amount of input lag, a trade‑off that developers will need to manage.
“AI‑generated frames are essentially a best‑guess of what the next frame should look like, and that guess can be remarkably close to reality when the model is well‑trained on gaming data.” – Digital Foundry analysis
What the Experts Are Saying
Mark Cerny, the architect behind both the PS5 and PS5 Pro, gave a concise preview:
“All I can say is that we have no more releases planned for this year. And that I look forward to discussing this more in the future.” – Mark Cerny, Digital Foundry interview
Game developer Emily Zhao of PixelForge Studios shared her perspective on integrating AI frame generation into a next‑gen title:
“When we tested early prototypes on a PS5 Pro with AI frame gen, we saw a 20 % boost in perceived smoothness without sacrificing visual fidelity. The key is to give artists control over the model’s confidence thresholds.” – Emily Zhao, Lead Technical Artist
Industry analyst Raj Patel from Future of Game Development noted:
“Sony’s move could force a paradigm shift, pushing other console makers to accelerate their own AI pipelines.” – Raj Patel
What This Means for Developers and Players
For developers, AI frame generation introduces a new set of tools and considerations:
- Performance budgeting: Teams can allocate GPU cycles to higher‑resolution textures while relying on AI to maintain frame rates.
- Artistic control: Adjustable confidence sliders let artists decide how aggressive the AI should be, preventing visual artifacts in stylized games.
- Testing pipelines: New QA processes are required to evaluate AI‑generated frames across diverse hardware configurations.
Players stand to gain immediate benefits:
- Higher perceived frame rates, especially in fast‑paced shooters and racing titles.
- Smoother VR experiences, where motion sickness is often linked to low frame rates.
- Potentially longer console lifespan, as AI can extend performance without hardware upgrades.
However, the technology is not a silver bullet. Games that already run at low native frame rates may see limited improvement, echoing the warnings from Nvidia and AMD about the need for a stable baseline.
How Sony’s AI Stacks Up Against Nvidia DLSS and AMD FSR3
| Feature | Sony AI Frame Gen | Nvidia DLSS 3 | AMD FSR3 |
|---|---|---|---|
| Core Method | Deep neural network predicts new frames | Neural network + optical flow | Interpolation (no ML) |
| Hardware Requirement | Dedicated AI accelerator (existing GPU) | RTX 40‑series Tensor cores | Any RDNA3 GPU |
| Latency Impact | Low‑to‑moderate (configurable) | Low (hardware‑accelerated) | None (pure interpolation) |
| Image Quality | High fidelity, especially in motion | Excellent, industry‑leading | Good, but can produce ghosting |
While Nvidia’s DLSS 3 remains the benchmark for AI frame generation on PC, Sony’s approach could level the playing field for console gamers, especially if the AI model is fine‑tuned for the unique constraints of console hardware.
Roadmap: When Will AI Frame Generation Arrive?
Based on Cerny’s comments and Sony’s product cadence, the likely timeline looks like this:
- Q4 2026: Developer preview kits for PS5 Pro with AI frame generation enabled.
- Early 2027: First wave of games (e.g., a major shooter and an open‑world RPG) ship with optional AI frame gen mode.
- Late 2027‑2028: Full integration into the PS6 launch, paired with Radiance Cores for ray‑traced AI rendering.
In parallel, Sony is expected to release SDK updates that let studios integrate the technology with minimal code changes, similar to how PSSR was rolled out.
Explore AI‑Powered Tools for Game Development
If you’re a developer looking to experiment with AI in your pipeline, UBOS offers a suite of ready‑made templates that can accelerate prototyping:
- AI Video Generator – create in‑game cutscenes with AI‑driven rendering.
- AI SEO Analyzer – optimize your game’s store page for discoverability.
- AI Chatbot template – build in‑game assistants that leverage the same models Sony will use for frame generation.
- AI Image Generator – generate concept art or texture variations on the fly.
Beyond templates, UBOS’s UBOS platform overview provides a low‑code environment for integrating AI services such as OpenAI ChatGPT integration or Chroma DB integration. These tools can help you prototype AI‑enhanced gameplay mechanics before the console SDK is publicly available.
Ready to future‑proof your studio? Join the UBOS partner program and gain early access to cutting‑edge AI modules that align with Sony’s upcoming frame‑generation pipeline.
Conclusion
AI‑generated frames represent a pivotal step toward truly fluid gaming on consoles. By leveraging deep learning to “imagine” missing frames, Sony aims to close the performance gap between console and high‑end PC gaming, while keeping the hardware cost in check. Developers will gain a powerful new lever for performance budgeting, and players can look forward to smoother, more immersive experiences across the PS5 Pro and the future PS6.
Stay tuned to UBOS for the latest templates, SDK updates, and expert insights that will help you harness this technology as soon as it lands on PlayStation hardware.
For the full technical interview with Mark Cerny, read the original Verge article.
Andrii Bidochko
CTO UBOS
Andrii Bidochko is an AI entrepreneur and researcher focused on AI agents, reinforcement learning, and autonomous systems. He writes about the technologies shaping the future of machine intelligence, from frontier models and agent architectures to real-world AI applications.