Google moves into AI coding helper battle with Codey and Studio Bot – Ars Technica

During today’s I/O show, Google announced Studio Bot, an AI assistant that Android developers can use to help write and debug code.

Built on Codey and the revised large language model PaLM 2, Studio Bot is only available to US developers at the moment and is in its “early days,” Google said. It’s part of Android Studio, Google’s official integrated development environment (IDE) for Android developers.

This differs from another Codey-based project and aims to compete directly with GitHub’s Copilot helper in inline code completion and generation.

While Copilot focuses on direct analysis of user code and providing inline suggestions, Studio Bot behaves similarly to Bard or ChatGPT in that it is a “conversational experience” that interacts with you as a kind of advisor. A promotional video depicting a Studio Bot advertising a developer asked “what is dark theme” (as if any self-developer wouldn’t know that) and then making follow-up queries to get code snippets to implement the dark theme in the developer’s program.

You don’t have to give Studio Bot any of your code to ask it for things. In fact, Google’s positioning suggests that, at least for the time being, it’s better suited to answering general API questions and the like than search and helping you build things directly.

Google warns against the developer documentation that “Studio Bot is still an early experiment, and may occasionally provide inaccurate, misleading, or false information while providing it with confidence. Studio Bot may give you working code that does not produce the expected output, or provide you with code that is not ideal or incomplete.”

This wasn’t the only developer-related I/O announcement. As mentioned briefly above, Google has also announced plans to launch a Codey-based code generator that is more akin to Copilot. Codey-based tools like this one work with JavaScript, Java, Python, SQL, and Go. This is part of a larger initiative to AI-ify nearly all of Google Cloud and other Google development tools and services over time.

Google also launched ML Hub, a repository of guidance for developers looking to train and use machine learning models in their work. A new experimental AI feature in the Play Store will allow developers to create a copy of app lists, analyze and summarize user reviews of apps, and more.