An April Fool's joke from nearly eight years ago has now evolved into a real tool for Nvidia.

The chipmaker on Tuesday unveiled an experimental version of Project G-Assist, an AI assistant designed to enhance the PC experience for gamers and creators. The feature is currently available to desktop users of GeForce RTX, its visual computing platform, through NVIDIA apps. The company said it will add broader support in the future.

G-Assist was first teased as a joke on Twitter in 2024 and later demonstrated at Computex in 0. It helps users manage key PC settings with simple voice or text commands, such as optimizing game and system performance, measuring frame rates, and controlling lighting.

Users can request real-time performance metrics, such as latency and temperature, and ask the system to optimize these metrics for specific games. According to the company, the assistant can also provide diagnostics and recommendations to reduce system bottlenecks, improve power efficiency, and overclock the GPU. It supports integration with third-party devices from companies such as Logitech G, Corsair, and Nanoleaf, allowing users to benchmark or adjust fan speeds, lighting, and other custom settings.

This comes as more and more companies are working to integrate human-like AI systems into their platforms, developing tools designed to handle everyday tasks and streamline digital experiences.

The launch of G-Assist also coincides with Nvidia's continued wave of next-generation technologies. At GPU Tech 2025, CEO Jensen Huang announced a series of tools designed to help humanoid robots learn how to be robots. The company also recently launched its latest line of AI chips and shared its evolving vision for quantum computing.

The beginning of an even bigger breakthrough

Julian Togelius, an associate professor of computer science and engineering at New York University who focuses on gaming and AI research, said Nvidia's latest efforts are a prime example of a useful domain-specific AI assistant.

"This shows that modern large language models have in-depth knowledge of many things, including computers, and can play a big role when getting the right information," he said. "At the same time, it reflects the complexity of configuring a gaming PC."

According to Togelius, this is just the beginning of similar accessibility features that are coming from Nvidia and other tech companies in gaming.