ARTEMIS by Google
ARTEMIS is an open source system that executes plain language requests on mobile devices by operating phone applications. It uses computer vision to navigate interfaces accurately.
The Blend
Google has publicly released ARTEMIS, an open source tool designed to let artificial intelligence control smartphone applications through standard text instructions. According to the project's GitHub documentation, the system uses visual recognition and automated interaction loops to navigate mobile screens much like a human operator.
For average mobile users, this represents a meaningful step toward truly capable virtual assistants that handle tedious phone chores. Rather than manually jumping between different services to map out travel routes and queue up videos, a user can provide a single instruction and allow the device to perform the actions autonomously. Tests on standard Android benchmark tasks showed a success rate exceeding 99 percent, according to the project maintainers.
However, it is still unclear how developers will safeguard these automated interactions against security risks, especially when agents access sensitive personal applications like mobile banking. Additionally, if app layouts frequently redesign their user interfaces, can visual AI models adapt instantly without triggering accidental purchases or misclicks?
Written independently by AI News Smoothie from the reporting listed below. Facts belong to the original publishers. Follow the links for their full coverage.
Ingredients
- GitHub - google/artemis: ARTEMIS turns natural-language instructions into reliable Android automation. It automates end-to-end workflows, captures logs, and integrates seamlessly with AI coding assistants such as Antigravity, Codex, and Claude Code. It also achieves 99%+ success rate on AndroidWorld Benchmark. · GitHub
Google published an open source project called ARTEMIS that allows AI software to control Android applications using natural language commands.