In principle the situation isn't that different from IPv4. A device on an IPv6 network can cycle through a series of random "temporary addresses" which it gener

In principle the situation isn't that different from IPv4. A device on an IPv6 network can cycle through a series of random "temporary addresses" which it generates itself, without contacting a DHCP server. Someone who gets your IPv6 address can track it down to a specific network, but there's no record of which device on that network initiated it, similar to IPv4+NAT. (A couple of caveats. One is that attackers can still correlate different connections within a short period of time, between temporary address rotations. Another is that temporary address support is broken on some Windows versions, so you may be leaking your MAC address all over the ace without knowing it.)