AI Select

Why It Exists

Users need to quickly extract and act on content visible on their screen without manual text selection or switching between apps.

How It Works

Galaxy AI Select uses Samsung's on-device vision AI to understand what is visible on the screen and intelligently select text, images, or content elements. The system leverages the NPU to perform real-time visual understanding, recognizing objects, text blocks, and content boundaries. Users can select content with a simple gesture and take action such as searching, translating, or editing the selected content.

Everyday Use Cases

• Screen content selection
• Image extraction
• Text grabbing
• Visual search
• Object identification
• Cross-app content sharing

User Workflow

• User activates AI Select via gesture or shortcut
• AI analyzes the current screen content in real-time
• User points at or circles the desired content
• System intelligently selects and extracts the content
• User chooses an action such as search, translate, or share

AI Processing Flow

The screen content is captured as an image and processed by on-device NPU vision models. The AI identifies text blocks, images, and objects using semantic segmentation. Selection boundaries are calculated based on content type and context. Processing happens entirely on-device for speed and privacy, with optional cloud enhancement for complex visual search queries.

Inputs / Outputs

Inputs:

• Screen capture
• Visual content
• Text blocks
• Image elements
• Object boundaries
• User selection gesture

Outputs:

• Extracted text
• Identified images
• Object labels
• Visual search results
• Translation options
• Share targets

Known Limitations

• May struggle with low-contrast text
• Image extraction quality depends on original resolution
• Works best with static content
• Some visual search queries require internet

Unsupported Scenarios

• Real-time video stream processing
• Handwriting recognition
• Non-Samsung app content
• Offline complex object recognition

Performance Notes

Visual analysis completes in under 300ms on NPU; Selection processing under 100ms; Cloud visual search 2-5 seconds; Battery impact negligible per use session

Available On

Shows where this feature is available and how its AI processing works on each platform. Availability may vary by device.

Research Status

Confidence and verification reflect how complete documentation is. Fields may show Not assessed or Not yet verified while research is ongoing - this flags gaps, not product deficiencies.

Research Notes

Available on Galaxy S24 series and newer; Requires One UI 6.1+; Works with any app displaying visual content; Introduced with Galaxy AI in 2024

Details

Capability Mapping

This feature maps to canonical capability family: Visual Understanding.

Canonical capability: Visual Understanding.

• Visual Understanding - Interprets what is in view and provides contextual understanding or actions.

Grounded in 6 device(s) on Galaxy AI.