Google is turning its CC artificial intelligence agent from a personal productivity tool into a household assistant, giving families a shared AI system that can coordinate calendars, email, tasks, and everyday responsibilities.

The search giant introduced the updated CC this week as part of a broader push to bring agentic AI into consumers’ daily lives. Unlike conventional chatbots that primarily respond to prompts, CC is designed to take actions on a family’s behalf, from organizing school schedules and appointments to preparing shopping lists and completing certain forms.

The move comes as AI companies increasingly look beyond workplace productivity and search toward the household as a potential market for autonomous digital assistants. Startups including Ollie and Fambot have also developed AI products aimed specifically at parents and families.

Google’s CC originally focused on personal productivity. It connected to Gmail, Google Calendar, Google Drive, and the wider web to understand a user’s schedule and deliver a “Your Day Ahead” briefing. Google later brought the feature into the Gemini app in May under the name “Daily Brief.”

The company said feedback from users showed that many wanted the technology to handle a broader set of household responsibilities, including children’s school schedules, sports practices, bills and meal planning.

Google has now repositioned CC around those needs.

A Shared AI Account for The Household

One of the biggest changes is that CC gets its own Google account, allowing it to collaborate with multiple members of a family while maintaining a separate set of permissions for the information it can access. Each family member determines what information they want to share. That could include emails about school events, sports teams, clubs, birthday invitations, medical appointments, or other household activities.

Users can forward relevant messages to CC’s email address or automate the process by selecting senders whose emails should automatically be shared with the agent. Parents could, for example, designate a school, sports club or travel company so that future messages from those organizations are automatically routed to CC.

The system can also suggest additional senders each week based on the emails it identifies as potentially relevant to the household. Google currently allows as many as six family members to use the system and collaborate with the AI agent.

Once information enters CC, the agent can turn it into actions. An email confirming an orthodontist appointment can result in an entry on the family’s shared calendar, while a school announcement about a teacher workday can automatically become a calendar event.

CC can also maintain shared tasks and track important dates, giving families a centralized system for managing responsibilities that would otherwise be spread across individual inboxes, calendars, and messaging threads.

The agent’s capabilities extend beyond organization.

Google says CC can fill out certain permission slips and activity registration PDFs, compile school-supply shopping lists, prepare weekly meal plans, calculate travel times between activities, and create shared Google Docs and Sheets.

When information is missing, the agent can ask family members for clarification and update its group memory so that it can use the information in future tasks. That represents a more consequential use of AI agents than simply generating text. CC is being positioned as a system that can interpret information arriving from multiple people, determine what needs to happen, and then execute selected tasks.

Google said the system runs on an isolated cloud computer powered by Gemini and the company’s agentic harness, Antigravity.

For now, however, the experiment has significant limits.

CC is available only to users in the United States who have personal Gmail accounts and are at least 18 years old. That excludes children and teenagers from directly participating in the system, even though they are often the source of many of the schedules, forms, and school communications that parents need to manage.

The restriction also means students cannot use school-provided email accounts with CC, even when those schools rely heavily on Google’s education products, Chromebooks and Google Workspace.

Existing CC users will receive invitations by email to upgrade their accounts, while new users can join Google’s waitlist.

AI Agents Move Into The Home

Google’s family-focused CC comes as competition around consumer AI agents intensifies. The appeal of these systems is their ability to move beyond answering questions. An agent can monitor information, maintain context, and perform actions based on a user’s instructions or established permissions.

Meta’s Muse personal AI agent is another example of the trend. Muse has risen to No. 1 among free applications on Apple’s U.S. App Store, ahead of ChatGPT, showing the growing consumer interest in AI systems positioned as assistants rather than conventional chatbots.

The family represents an especially complicated environment for such technology because responsibilities, information, and permissions are shared across multiple people.

A household AI agent needs to know which information belongs to whom, what can be shared with other family members, and which actions it is authorized to take. Google’s decision to give CC its own account and separate permissions appears designed around that problem.

It also creates a different relationship between users and AI. Rather than asking a chatbot a question when needed, family members can potentially allow an agent to remain embedded in their everyday administrative routines. That could make household management a significant testing ground for agentic AI. But it also means that questions around access, privacy, authorization, and mistakes become more important as agents gain the ability to act on information rather than simply provide answers.

Chinese AI startup Manus is in talks to raise about $500 million at a valuation of roughly $4 billion, according to The Wall Street Journal, as the company rebuilds its business as an independent company after Beijing blocked a proposed $2 billion acquisition by Meta.

The fundraising discussions would give Manus a significantly higher valuation than the roughly $2 billion level reportedly used by early investors and backers to buy back shares as the startup separated itself from the U.S. social media giant.

Potential investors in the new round include IDG Capital, Boyu Capital and battery manufacturer Contemporary Amperex Technology, alongside existing backers Tencent, HSG and Zhenfund, the Journal reported, citing people familiar with the matter.

Manus is also considering a corporate restructuring that could prepare the company for a potential initial public offering in Hong Kong, according to the report.

The fundraising effort marks a rapid change in the startup’s trajectory. Manus became one of China’s most closely watched AI startups after an AI agent demonstration went viral last year, drawing attention to its ability to perform tasks that traditionally required users to work across multiple software applications.

The company later moved its staff to Singapore in mid-2025 and agreed to a $2 billion acquisition by Meta in December.

At the time, Manus was reportedly generating more than $100 million in annual recurring revenue, giving Meta a rapidly growing AI business and a team working on agentic software that could complement its broader artificial intelligence ambitions.

The deal ultimately became entangled in a much wider debate in China over the loss of AI researchers, engineers and technology companies to the West.

From Meta Acquisition to Independence

Beijing eventually blocked the transaction, citing potential violations involving export controls and foreign investment rules.

The intervention highlighted the complicated environment for Chinese AI companies with international operations and Western investors. A transaction that might ordinarily have been assessed primarily on its commercial value became linked to concerns about technology transfer, national interests and the movement of Chinese AI capabilities outside the country.

Manus has since been unwinding its relationship with Meta.

The company’s early investors and backers reportedly helped facilitate a share buyback at a valuation of about $2 billion, allowing the startup to re-establish itself outside Meta’s ownership.

In August, Manus told users that they would need to export and back up their own data because information generated after Meta’s acquisition would have to be deleted to “comply with regulatory requirements in specific jurisdictions.”

The company said this month that it had resumed independent operations and that its founding team would continue to lead the business. The potential $500 million financing would therefore represent more than a conventional growth round. It would provide capital for a company that has had to reconstruct its ownership structure and operating model after a major cross-border transaction collapsed.

It would also give Manus a substantially higher valuation if completed at the reported $4 billion level.

But that valuation is expected to depend heavily on Manus’ ability to demonstrate that its early viral momentum has translated into durable revenue growth and a defensible position in the increasingly crowded AI-agent market.

Manus Targets the Agentic AI Market

Manus operates in a segment of AI that has attracted significant investment as companies move beyond conventional chatbots toward systems capable of completing multi-step tasks.

Its products include a chatbot and AI-powered coding tools that allow users to build applications and websites, create designs and presentations, generate video, and interact with the web through a browser assistant.

The company’s offerings overlap with products being developed by OpenAI and software-focused AI companies such as Lovable and Replit. That puts Manus in a market where the underlying technology is advancing quickly, but competitive barriers remain difficult to establish.

AI agents are now being positioned as interfaces through which users can delegate tasks rather than simply ask questions. The commercial opportunity is potentially large, but companies must still demonstrate that agents can complete tasks reliably enough for users to trust them with more consequential work.

Manus’ reported revenue trajectory suggests it had already established meaningful commercial demand before the Meta transaction. The challenge now is to convert that momentum into an independent business while competing against much larger AI companies with significantly greater access to computing resources and capital.

The reported Hong Kong IPO preparations add another dimension.

A listing in Hong Kong could provide Manus with access to public-market capital while keeping its corporate future more closely tied to China’s financial system. It would also allow investors to assess the economics of an AI-agent company whose ownership and international operations have already been shaped by geopolitical restrictions.

The potential IPO is not yet a commitment, and neither the reported funding round nor the restructuring appears to have been finalized.

Still, the sequence of events underpins how China’s AI industry is increasingly being shaped by forces beyond product development. Manus went from a viral AI startup to a proposed acquisition target for Meta, then became caught in China’s effort to retain control over strategic technology and talent, and is now attempting to establish itself as an independent company again.

If the reported financing proceeds at $4 billion, Manus would be asking investors to value the company not on the failed Meta transaction, but on its prospects as an independent AI-agent platform. That makes the next stage especially important. The company must show that its technology can generate recurring demand, that its products can compete as AI agents become increasingly commoditized, and that its corporate structure can navigate the regulatory constraints that helped derail its previous path.