Money in exchange for code to train AI: Google ready to pay Android developers for source code

Training AI models requires a vast amount of data and, above all, involves a constant update of the database with the need to "feed" the AI new data for training. A new report once again confirms that companies active in the AI sector are willing to spend a lot of money to obtain the data to train their models.

Is Google offering money to Android app developers?
From information that has emerged in recent hours, thanks to a report by 404 Media, Google has launched a pilot project involving some Android app developers who are invited to share the source code of their applications. The goal is to allow AI models (specialized in programming) to train and improve by utilizing new data, such as the complete code of functioning applications on Android. According to Google, this program represents an additional way to monetize the existing code while maintaining copyright.

Additionally, developers will not have exclusivity constraints and can share their code with other companies for further monetization opportunities. Google has also published a web page inviting developers, professionals from various sectors, and creators to share their work for AI training.

In the email sent by Google to developers, it states:
"Receive compensation for sharing the source code of your apps, as well as your archived projects. Whether it is the active source code of your current app or archives of prototypes and secondary projects no longer in use, that code could hold unexpressed value. This is a unique opportunity to help transform tools and products, support the developer ecosystem, and generate new revenue."
The 404 Media report clarifies that the text does not explicitly reference artificial intelligence, although it contains a link to the previously mentioned web page where the importance of improving AI-based products through data sharing is emphasized. At the moment, Google has not provided comments on the matter. However, the company may need a program like this to improve the performance of its AI in code generation, a field where other models may currently offer superior performance.