Browser developer Opera has rolled out a novel web standard—WebAI—that empowers users to execute AI models straight in the browser, eliminating the need for cloud service dependence. WebAI is tailored for a diverse range of open-weight AI systems; it is not restricted to a specific model and is fully compatible with open-weight models like Google Gemma. It also facilitates the utilization of Apple's on-device foundational models on Mac devices. Presently, the WebAI functionality is accessible as an early preview in the Opera One browser and will be progressively introduced to Opera GX and Opera Neon, with users needing to activate it manually. WebAI enables developers to integrate AI capabilities into their products without the necessity of deploying AI infrastructure. Simultaneously, users can sidestep the need for subscribing to cloud-based AI services since data processing transpires locally, thereby aiding in privacy preservation. The feature accommodates operations such as text generation, content classification, and the organization of information relationships within the browser. Nonetheless, WebAI's real-world performance is contingent upon the hardware configuration of the user's device, as it harnesses local computing resources during operation, which could potentially escalate hardware load and impact device performance.