People must consent to run exit nodes, as running exit nodes puts them at more of a legal risk than just running a relay node that passes traffic. if you’re accessing an unencrypted website, the exit node can potentially monitor your Internet activity, keeping track of the web pages you visit, searches you perform, and messages you send.
If you’re accessing an encrypted (HTTPS) website such as your Gmail account, this is okay – although the exit node can see that you’re connecting to Gmail. In the below diagram, the red arrow represents the unencrypted traffic between the exit node and “Bob,” a computer on the Internet. This node where traffic exits the Tor network is known as an “exit node” or “exit relay.” The last Tor node, where your traffic leaves the Tor network and enters the open Internet, can be monitored. For example, let’s say you are connecting to Google through Tor – your traffic is passed through several Tor relays, but it must eventually emerge from the Tor network and connect to Google’s servers. However, most Tor traffic must eventually emerge from the Tor network. Your computer may have initiated the connection or it may just be acting as a relay, relaying that encrypted traffic to another Tor node. Tor is designed so that it is theoretically impossible to know which computer actually requested the traffic. In summary, when you use Tor, your Internet traffic is routed through Tor’s network and goes through several randomly selected relays before exiting the Tor network. Read our discussion of how Tor works for a more detailed look at how Tor provides its anonymity.