AI agents that browse the web are not new. HUMAN tracked a 7800% increase in agentic traffic in 2025 and consistent traffic gains throughout 2026. Early adopters experimented with tools like ChatGPT Agent, Claude for Chrome, and Perplexity Comet.
Meta’s recently released Muse agent signals a larger and faster shift in how people interact with the internet: the first mass-market adoption of agentic AI.
Muse lowers the barrier to adoption. Most other AI agents require a paid subscription for browser automation. Muse has a generous token limit on the free tier, includes an AI browser agent by default, and is designed for everyday consumers.
How Muse changes the equation for agentic traffic
Muse comes from a company billions of people already have an account with, giving Meta a vast distribution network. Facebook, Instagram, Messenger, and WhatsApp give Muse an ability to scale that OpenAI and Anthropic lack.
By September 21st, Muse had reportedly exceeded 2.5 million downloads and reached the top of U.S. app charts. 95% of Muse’s users are also Facebook users, while 63% are Instagram users, according to Apptopia.
Muse will be many users’ first experience of an agent browsing and acting on their behalf.
The consequential signal for businesses will be the scale of this new agent’s traffic, where the sessions actually go, and what they do.
What HUMAN is seeing
In HUMAN’s early observations, Muse accounts for an average of 72% of daily AI-agent requests. Muse’s daily request volume is 5.2 times that of the next largest, ChatGPT Agent.
By September 27th, daily Muse requests outnumbered requests from all other measured agents by a factor of 2.34×.
To put that in context: Muse reached a request volume that took ChatGPT 11 months to achieve–in just two days.
Among Muse requests, 84% target product and search pages.
Like other agents, Muse is moving through the same discovery-to-purchase paths that businesses measure, optimize, and protect. In a recent sample, Muse is 25% more likely to reach checkout pages than other AI agents. An agent researching products for a real shopper may be valuable. An agent that spoofs its identity, abuses an account, or attempts a fraudulent transaction is not. Counting both as undifferentiated “bot traffic,” or treating an agent session as a human session, can distort conversion rates, funnel performance, attribution, and fraud signals.
From visibility to governance
Given the velocity in the space, it’s no surprise that 96% of CISOs cannot accurately quantify agentic traffic. Emerging agentic traffic sources can contaminate the data that businesses and technical teams rely on to support policy decisions. The immediate need is accurate, real-time identification of this traffic, followed by governance that connects business policy to technical enforcement.
There has been work on internet standards, particularly web-bot-auth, which HUMAN supports, to identify AI agents. But most AI providers do not yet use these standards. Organizations increasingly need their agentic strategy to account for how agents navigate customer journeys and affect capacity, conversion, and data quality.
They also need to find the right balance between allowing trustworthy agents, blocking risky actions, and evaluating what activity drives net value for the business.
Activity determines policy
Today, through AgenticTrust, HUMAN identifies and verifies Muse and more than 30 other AI agents using proprietary research or digital signatures that cannot be spoofed. With comprehensive visibility into how humans, bots, and AI agents interact with their digital properties, organizations can enforce their company’s AI policies: allowing the automation that creates value and blocking activity that creates risk. This includes granular control of whether actions should be allowed or blocked, such as add to cart or log in, or rules on which AI agent is trusted.
The first step to addressing this traffic is visibility. If you want to understand what Muse and other AI agents are doing across your business, AgenticTrust gives you the visibility to separate useful activity from noise. See how it helps teams measure business impact, manage access, and make smarter decisions about agentic traffic with a demo today.
Want to know more about Muse? Join our webinar on October 1st to learn: