Meta's Muse AI Agent Integrates with Instacart for Grocery Orders

Meta has launched Muse, an artificial intelligence agent that integrates with Instacart to handle grocery shopping tasks, according to a Fox Business report. The technology represents Meta's latest push into practical AI applications that could reshape how consumers interact with everyday services.

The AI agent demonstrated strong performance during its development phase, achieving a 95% task success rate in testing that included human concierge evaluations. This high completion rate suggests Muse can reliably interpret shopping requests, navigate product selections, and execute orders through the Instacart platform.

Mixed reception from investors

Despite the technical achievement, the announcement has not uniformly impressed Wall Street. Investment strategist Brian Belski stated his firm remains unwilling to buy Meta stock even with the AI agent's demonstrated capabilities. The skepticism highlights a broader tension in technology investing: strong product execution does not always translate to immediate shareholder enthusiasm, particularly when questions remain about monetization paths and competitive positioning.

The partnership with Instacart gives Meta's AI a direct connection to a grocery delivery infrastructure serving millions of households. By handling the cognitive work of meal planning and shopping list management, Muse could reduce friction in a routine task that many consumers find time-consuming.

Why it matters

Meta's move into AI-powered shopping assistants signals how large technology platforms are racing to embed artificial intelligence into transactional services. Success in this space could create new revenue streams and deepen user engagement, but it also puts Meta in direct competition with Amazon's shopping capabilities and other e-commerce players investing heavily in AI. The cautious investor response underscores that technical milestones alone may not be enough—companies must demonstrate clear paths to profitability and sustainable competitive advantages in crowded AI markets.

Testing approach and capabilities

Meta employed human concierge testing as part of its validation process, comparing Muse's performance against human shoppers to establish reliability benchmarks. This methodology provides a concrete measure of whether the AI can match human judgment in understanding preferences, making substitutions, and handling the nuances of grocery selection.

The technology's capabilities extend beyond simple list fulfillment. AI agents like Muse are designed to understand natural language requests, learn user preferences over time, and adapt to changing inventory and pricing conditions—functions that require sophisticated language models and decision-making algorithms.

Details about Muse were first reported by Fox Business during a segment on the Varney & Co. program that aired September 23, 2026.