● Covered by 1 source · 2 reportsMedium impact2 neutral
Cloudflare introduces new analytics and policies for personal agent traffic
Key points
- New analytics categorize personal agents in Security Analytics.
- Personal Assistants widget shows specific agents and request counts.
- Site owners can filter traffic details by individual agents.
- Account Abuse Protection integrates agent activity with user accounts.
- AI agents like Meta's Muse and OpenAI's dots act on behalf of human users.
- Agentic traffic may soon surpass human browsing.
- Sites need new mechanisms for identity, reputation, and accountability for agents.
- Personal AI Agents are automations taking instructions from a human to pursue a goal.
Enhanced Visibility for Personal Agents
Cloudflare has introduced new analytics features to help site owners identify and monitor personal agent traffic. Previously, interactions from personal agents, such as product comparisons or purchase completions, appeared as undifferentiated automated traffic. The new tools provide a clear view of which agents are visiting a site and their activities.
Detailed Agent Tracking in Security Analytics
In Cloudflare's Security Analytics, the Bot Analysis chart now includes 'Personal Assistants' as a distinct category, alongside 'Verified Bot', 'Likely Human', 'Likely Automated', and 'Automated'. This allows security teams to isolate and analyze personal agent traffic over time. A new 'Personal Assistants' widget in Top statistics lists specific agents accessing the site and the number of requests each sends.
Granular Control and Filtering
Site owners can now filter all other widgets in Security Analytics by a specific personal agent. This provides detailed insights into the paths an agent requests, the status codes it receives, and its network origins. This functionality helps determine if an agent is performing expected actions like logging in, browsing product pages, or reaching checkout, or encountering errors.
Integration with Account Abuse Protection
For sites utilizing Account Abuse Protection, site owners and fraud teams can now see which user each personal agent represents. The population overview page displays active personal agents across user accounts and their usage frequency. This integration also shows if a personal agent made a login or sign-up request, enabling fraud teams to filter accounts by AI assistant activity.
Updates
- AI agents like Meta's Muse and OpenAI's dots act on behalf of human users.
- Agentic traffic may soon surpass human browsing.
- Sites need new mechanisms for identity, reputation, and accountability for agents.
- Personal AI Agents are automations taking instructions from a human to pursue a goal.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Primary sources
CVE CVE-2023-503877.5 HIGHHow outlets covered it
AI agents, such as Meta's Muse and OpenAI's dots, are increasingly acting on behalf of human users on websites, creating challenges for sites in identifying and holding users accountable. These agents can perform actions without human limitations, leading to a potential shift where agentic traffic may soon surpass human browsing. Websites need new mechanisms to preserve the relationship between sites and users, focusing on identity, reputation, and accountability, as traditional blocking of intermediary traffic becomes unfeasible.
Cloudflare launched new analytics and granular access policies to provide site owners with better visibility and control over personal agent traffic. These tools allow businesses to understand how personal agents interact with their sites and manage access based on specific needs, rather than blocking all automated traffic.