More from this creator
Other episodes by Kitty Cat.
More like this
If you liked this, try these.
Transcript
The full episode, in writing.
Meta Platforms is developing a consumer AI agent tool currently known as Hatch. Internal documents from April and May 2026 indicate Meta is considering charging up to $200 per month for a premium version of Hatch. The premium tier, as revealed in reports from June 2026, is positioned to directly compete with high-end AI subscriptions like OpenAI’s ChatGPT Pro and Anthropic’s Claude Max, both of which are priced at $200 per month. Charging for a premium tier puts Meta’s AI services in direct competition with other major players in the rapidly growing agentic AI market.
Hatch is designed to automate a range of tasks for users, including generating custom software tools, managing schedules, sending emails, and streamlining workflows based on plain language instructions. Users will be able to ask Hatch to create new applications, such as a fitness tracker, by simply describing their needs, and the agent will generate a working tool. This approach aims to make the process of coding and automation accessible to people without a technical background. The product is conceived as a more user-friendly, consumer-focused version of the open-source OpenClaw agent tool, which is known for its flexibility but can be complicated for non-expert users.
The interface for Hatch will include a customizable dashboard where users can access and manage the tools and automations created by the agent. This dashboard is similar to OpenClaw’s “skills” system—a modular framework where users can select from different add-ons that allow agents to automate workflows and interact with third-party services. For example, one skill could help create a travel itinerary, while another might automate the process of booking appointments or managing emails. The modular design enables users to build highly personalized workflows without needing to write code, lowering the barrier to entry for advanced automation.
A premium subscription tier, potentially called “Hatch Plus,” is under active consideration. This tier would offer between five and ten times higher usage limits than the free version of Hatch, according to internal planning documents. Usage is measured by token allowances, a system common among AI subscription products. These tokens represent the computational resources or number of requests a user can make each cycle, and allowances would reset each billing period without rolling over to the next month. This structure mirrors the business models of other leading AI agents, ensuring that heavy users pay more for higher capacity while casual users can access the basic service for free or at a lower price.
During its development, Hatch has been powered by Anthropic’s Claude Opus 4.6 and Claude Sonnet 4.6 models. However, the final product is expected to launch using Meta’s own AI model, Muse Spark. Muse Spark is Meta’s latest generation model, developed in-house as part of the company’s broad investment in AI infrastructure and research. By transitioning to Muse Spark, Meta aims to reduce reliance on third-party providers, improve performance, and better control the integration of new features. This shift also allows Meta to leverage its extensive compute resources and proprietary data to drive differentiation in the consumer AI market.
Meta targeted a U.S. launch for Hatch in April 2026, with an initial pilot rollout to about 10 businesses for early testing. According to an internal memo from early May, the plan called for a broader launch with finalized branding and product naming in July 2026, though launch dates remain subject to change. As of the most recent reporting, internal testing was scheduled to begin by the end of June 2026, but no public release date has been announced. Documents from May also discuss ongoing deliberations over the product’s final name and branding, showing that even core aspects like the “Hatch” name were not yet confirmed.
Meta’s move into paid AI subscriptions for both consumers and businesses is a core part of its strategy to diversify revenue beyond advertising, which remains its primary income source. In 2025, Meta’s AI-driven revenue exceeded $200 billion, with almost all of that still coming from advertising. The company’s capital expenditures for AI infrastructure and model development are projected to reach $145 billion in 2026, about double the level of 2025, putting significant pressure on Meta to develop new, sustainable revenue streams.
Mark Zuckerberg has outlined a vision for building AI agents that serve both business and individual users, with the long-term goal of achieving what he calls “personal superintelligence.” On an earnings call in April, Zuckerberg described Meta’s ambition as not just delivering AI assistants, but building agents that understand user goals and proactively work to help users achieve them around the clock. He has also highlighted the potential for users to “vibe-code” their own tools—using natural language to describe an idea and letting the agent do the technical work to bring it to life. Zuckerberg expects this approach to unlock a surge in entrepreneurship and innovation by giving people tools that previously required specialized skills.
Meta has already taken steps towards monetizing its AI products. The company has rolled out paid features for its Meta AI chatbot, which is accessible through Facebook, Instagram, and WhatsApp. This includes subscription plans for features like Facebook Plus, Instagram Plus, and WhatsApp Plus, announced in May 2026, which offer higher usage limits for the AI chatbot. According to Meta’s Head of Product Naomi Gleit, these paid tiers are part of a broader push to offer premium AI features to both consumers and businesses across Meta’s platforms.
Alongside Hatch, Meta is also developing AI-powered shopping tools for Instagram, targeting a launch before the end of 2026. These tools are intended to help brands and creators with analytics, engagement management, and customer activity tracking. For business users, Meta is testing new subscription plans that provide brand management and analytics, allowing businesses to automate customer service, recommend products, close sales, and book appointments through Meta’s platforms. The Meta Business Agent tools, currently free, are expected to transition to a paid subscription model in the coming months.
To train and test Hatch, Meta has built sandboxed simulated web environments that include popular platforms like DoorDash, Etsy, Reddit, Yelp, and Outlook. These environments allow Hatch to practice interacting with real-world platforms, improving its ability to automate tasks like shopping, booking services, or managing communications across different sites. This behind-the-scenes infrastructure investment is intended to ensure Hatch can handle a wide range of user requests across the digital landscape, not just within Meta’s own ecosystem.
As of June 2026, Meta is reviewing tiered pricing options for Hatch, with the premium subscription pricing at $199.99 per month under serious consideration. The final structure and rates have not been announced, but internal memos confirm that higher tiers would include substantially more computational capacity and priority access. This pricing targets the same segment as the highest-tier plans offered by OpenAI and Anthropic, positioning Meta as a direct competitor in the market for advanced AI agents.
Meta’s AI product expansion is also driven by investor expectations for new revenue streams. With nearly all of Meta’s 2025 revenue—more than $200 billion—still dependent on advertising, the company is under pressure to justify unprecedented capital spending on AI model research, infrastructure, and chip development. By commercializing advanced agentic tools like Hatch, Meta aims to convert these investments into recurring, high-margin subscription revenue.
Meta’s efforts reflect a wider industry trend, as tech giants increasingly seek to monetize proprietary AI technologies through consumer and business subscriptions. The race to deliver agentic AI for everyday tasks is reshaping digital services and expanding the definition of what personal computing can do. The latest development the sources cover is Meta’s internal review of a $199.99 monthly premium subscription for Hatch, with public release plans still in flux and the underlying technical foundation shifting to Meta’s own Muse Spark AI model.