Microsoft has announced the introduction of new APIs for its Edge web browser, enabling developers to integrate AI capabilities directly into their web applications. This reveal came during the Build 2025 conference and is seen as a strategic move to compete with Google Chrome, which also offers built-in AI functionalities for developers.
Among the key features of these APIs is access to the Phi 4 mini AI model, which was launched by Microsoft in February and boasts an impressive 3.8 billion parameters. This model has been specifically designed to excel in problem-solving, particularly for mathematical computations, thanks to its training data that includes both human-generated and synthetic datasets. Generally, a higher number of parameters equates to better performance, yet Phi 4 mini’s smaller size allows for more efficient processing on devices with limited computing power, like laptops and mobile phones.
Further expansions in Edge will include additional AI APIs, notably a set designed for writing assistance that will help in generating, summarising, and editing text. Microsoft is also set to introduce a translator API within a few months, aimed at supporting AI-driven translations for Edge users.
Microsoft has described these experimental APIs as potential web standards, ensuring compatibility across different platforms and browsers, as well as with other AI models. This is particularly significant for developers operating in environments where sensitive data is involved, as the APIs provide a level of privacy and security by processing information directly on devices, rather than routing it through cloud services.
The new AI capabilities are currently accessible through Edge’s Canary and Dev channels, allowing developers early access to test and implement them in their projects.
In addition, Microsoft is enhancing Edge’s PDF features, introducing a translation tool that supports over 70 languages. This feature allows users to open a PDF file in Edge, click on a translation icon, and create a new document in their preferred language. The PDF translation tool will become generally available next month, with Canary users able to experiment with it starting from now.
Overall, Microsoft’s initiative to extend the capabilities of Edge with these new AI features reflects its commitment to advancing web application development, while also prioritising user privacy and device efficiency.
Fanpage: TechArena.au
Watch more about AI – Artificial Intelligence
