Snap Launches Specs Intelligence AI Assistant on iOS
Snap has introduced Specs Intelligence, an anticipatory AI assistant designed to connect user accounts and proactively manage daily tasks across iOS, Mac, and its new AR glasses.

Snap has officially unveiled Specs Intelligence, a proactive artificial intelligence assistant designed to integrate with a user's digital accounts to streamline daily workflows. The technology is launching alongside Specs, the company's inaugural consumer augmented reality glasses. Beyond the wearable hardware, Snap is rolling out an iOS preview of the assistant, while Mac users can join a waitlist to access the complete early-access version.
The system functions by analyzing user habits, schedules, and relationships across connected applications such as Gmail and Slack. Unlike traditional reactive chatbots, Specs Intelligence is built to anticipate user needs and surface relevant information without requiring direct prompts. For example, the assistant can compile travel itineraries, highlight key discussion points before a meeting, or flag upcoming work deadlines that conflict with personal plans. When paired with the Specs AR glasses, the assistant projects this contextual data directly into the user's line of sight.
According to Snap spokesperson Cassie Bumgarner, the assistant relies on a "proprietary combination of open-source models hosted in the US alongside local LLMs." To address the privacy concerns inherent in granting an AI access to personal communications, Snap states that it cannot view or access the personal content processed by the tool. Furthermore, the company promises that data from connected accounts will not be used to train its models or target personalized advertisements.
For developers and tech practitioners, Snap's approach represents a significant push toward ambient computing, where AI operates continuously in the background rather than waiting for explicit user queries. By combining cloud-based open-source models with on-device local LLMs, Snap is attempting to balance the high computational demands of proactive assistance with the strict latency and privacy requirements of wearable AR hardware.
This is our own summary of reporting by The Verge AI


