Meta is making a comprehensive bet on AI agent technology, viewing it as the next generation of core product form, and is advancing two key strategic fronts internally.
According to a May 5th report by The Information, citing people familiar with the matter, Meta is developing a consumer-facing AI agent product, internally codenamed "Hatch," with the goal of completing internal testing by the end of June. The company also plans to integrate a standalone AI shopping agent tool into Instagram, aiming for a launch before the fourth quarter of this year.
The report states that these initiatives are a direct reflection of CEO Mark Zuckerberg's push for an AI agent strategy. He stated during the company's quarterly earnings call last week that Meta's goal is to "build agents that can understand user goals and help users achieve those goals around the clock."
Concurrently, Meta announced it is increasing its capital expenditure for AI infrastructure this year to a maximum of $145 billion. The launch of an Instagram shopping agent is seen internally as a crucial asset for direct competition with TikTok Shop.
**Hatch: From OpenClaw to In-House Development** The report indicates that the development of Hatch was directly inspired by OpenClaw—a tool that gained rapid popularity in tech circles for its ability to build autonomous AI agents. Zuckerberg had attempted to acquire OpenClaw, but according to its founder, Peter Steinberg, on a podcast, the tool was ultimately acquired by OpenAI in February, leaving Meta's acquisition attempt unsuccessful.
In this context, Meta shifted to independently developing a consumer-grade agent with similar capabilities. Reportedly, Hatch is currently powered by Anthropic's Claude Opus 4.6 and Claude Sonnet 4.6 models, but will transition to Meta's latest in-house AI model, Muse Spark, upon its official launch.
To test Hatch's practical capabilities, Meta has specifically built a "sandbox" network environment that simulates real websites like DoorDash, Etsy, Reddit, Yelp, and Outlook, allowing the agent to train and test under controlled conditions.
During the earnings call, Zuckerberg also acknowledged the technical challenges of bringing an OpenClaw-like tool to Meta's billions of users, including high dependence on massive-scale infrastructure and ensuring the product is simple enough to use—OpenClaw remains overly complex for most non-technical users.
**Capability Iteration: Proactivity, Memory, and Tool Usage** Meta is currently accelerating the enhancement of Hatch's core capabilities across multiple dimensions.
According to the report, the company is focusing on boosting Hatch's proactive decision-making ability, enabling it to take autonomous actions at appropriate times rather than passively waiting for user commands.
Simultaneously, Meta is expanding the model's context window and strengthening its cross-conversation memory, allowing the agent to persistently remember relevant user information across different sessions. Additionally, the company is optimizing the agent's response methods and the logic it uses for selecting and calling external tools.
It is noteworthy that Meta already has an internal AI agent for employee use, called MyClaw, which is used to access work documents, summarize internal company forum posts, and obtain technical advice.
However, as previously reported by The Information, MyClaw recently triggered a significant security alert within Meta—an employee followed incorrect advice from the agent, leading to sensitive company and user data being accessed by unauthorized personnel. This incident highlights the ongoing real-world challenges regarding the reliability of AI agents.
**Instagram Shopping Agent: A Direct Challenge to TikTok Shop** On the consumer front, Meta's other strategic front involves deeply embedding an AI shopping agent into Instagram.
According to the report, the core functionalities of this tool include allowing users to directly click on products within Instagram Reels or their feed to view detailed information, navigate to external web pages, and complete purchases within the platform without leaving the app.
This functionality will build upon the enhanced AI shopping experience Meta released in March, which included presenting richer product information via AI and a new checkout process allowing users to complete purchases by clicking directly on ads.
Sources also revealed that Meta hopes this tool will increase competitive pressure on TikTok Shop. TikTok Shop has established a significant advantage in social commerce through the deep integration of short-form video and e-commerce, and Meta's move is intended to counter this using Instagram's vast user base and AI capabilities.
In the broader competitive landscape, Alphabet launched Gemini Enterprise for Customer Experience in January, which includes AI shopping agent features supporting product recommendations, shopping cart building, and automated checkout with user authorization. Amazon.com offers its AI shopping assistant, Rufus, to help users track prices, research, and purchase products.
Analysis suggests that the advancement of both Hatch and the Instagram shopping agent epitomizes Meta's search for commercial returns on its massive capital investments in AI. Zuckerberg has positioned AI agents as the core of his vision for "personal super-intelligence," describing them on the earnings call as all-day tools to help users achieve their goals.
As Meta raises its AI infrastructure capital expenditure ceiling to $145 billion this year, market attention is intensifying regarding when these AI investments will translate into substantial revenue.
Whether Hatch can find a balance between technical complexity and mass usability, and whether the Instagram shopping agent can effectively leverage the social commerce market, will be key indicators for assessing the success of this strategy.