Google has announced the launch of a suite of new AI-powered tools within the AdMob platform, aimed at enabling app developers to manage monetization processes and develop their applications with higher efficiency. These tools include the intelligent assistant Ask AdMob and Agent skills, in addition to the development of an MCP server to connect development environments directly to AdMob data, reducing the manual effort required to manage advertisements.
Ask AdMob Intelligent Assistant
The "Ask AdMob" assistant is based on the Gemini model to provide an interactive experience within the user's account, acting as an expert in monetization strategies. This tool allows developers to perform several vital tasks through natural language commands, including:
- Fixing setup issues: Detecting and analyzing Software Development Kit (SDK) integration errors and addressing any malfunctions affecting ad delivery.
- Revenue optimization: Identifying new profit opportunities by comparing app performance with peer benchmarks and suggesting high-value bidders to optimize mediation groups.
- Brand safety adjustment: Setting up custom rules to protect brand safety using natural language, instead of complex manual management of blocklists.
Programming Automation via Agent Skills and MCP Server
Google is moving to a more advanced stage through Agent skills, which are ready-to-use playbooks that allow intelligent agents to execute specific programming tasks, such as migrating to new SDK versions or validating the implementation of ad code within the app.
To enhance this integration, Google is developing a Model Context Protocol (MCP) server that will allow intelligent agents to have direct access to developers' coding environments. This development will enable agents to read account data, create new ad units, and manage mediation settings automatically, reducing the need for constant navigation between dashboards and codebases.
What changes practically for developers?
This approach represents a shift from manual dashboard management to what is known as an "Agentic workflow." Instead of spending hours copying and pasting ad unit IDs or manually searching for integration errors, developers will be able to direct complex technical tasks through conversation. However, Google emphasized the principle of human control; all recommendations and actions suggested by the AI will require explicit review and approval from the developer before being applied to live applications.
Timeline and Future Outlook
Testing for these features will begin over the next few months, with a plan to expand the capabilities of intelligent agents and introduce more advanced tools during 2027. Through this step, Google aims to reduce routine operational burdens, allowing developers to focus more on product development and improving user experience.