Potential Unblinding Attack through Deterministic ID Blinding Recalculation by Resourceful Attacker with a Large Number of Service Nodes

[venezuela01]

According to swarm id coverage of individual service node operator, OPTF currently controls 50% of swarm IDs, despite only operating 10% of the nodes. The second-largest operator, L6qq, controls approximately 36% of swarm IDs, yet only operates around 7.5% of nodes.

Given this high coverage of swarm IDs, large operators can gain access to a significant portion of user Session IDs. If an attacker obtains a substantial number of Session IDs, they can replicate the deterministic ID blinding algorithm in PySOGs, thereby linking the blinded ID back to its original, unblinded state.

In a worst-case scenario, if the attacker simultaneously operates a large number of service nodes and PySOG servers, they would be able to uncover a significant number of unblinded IDs.

The next most severe scenario would involve an attacker who controls a substantial number of service nodes and subscribes to many PySOG communities. This could expose a large portion of users who have posted in these communities, potentially revealing their unblinded IDs.

It's crucial to note that this doesn't necessarily increase the risk of spam, as attackers operating a high number of nodes would already have knowledge of unblinded IDs. The principal concern lies in the potential exposure of a user's interests of certain type of contents or topics, which could be linked to their unblinded ID.

While the damage from such an attack may be limited, it's essential to clearly communicate these risks to our users. In this way, they can make informed decisions.