The going dark argument is still wrong after 15 years

In 2011, law enforcement agencies warned that end-to-end encryption would make criminal investigations impossible. They called it going dark: a future where encrypted messages put criminals beyond reach. Fifteen years later, a new paper revisits that prediction. WhatsApp encrypts two billion users by default. Signal exists. iMessage locks down conversations automatically. The going dark future arrived. Except investigations did not stop. Prosecutors still build cases. The paper tracks how: metadata, cloud backups that skip encryption, seized devices with unlocked sessions, informants who screenshot chats. Encryption protected the pipe, not the endpoints. The technical argument was always narrow. E2EE stops wiretaps on the network. It does not stop lawful device searches, does not stop compromised clients, does not stop users who paste secrets into unencrypted notes apps. Most arrests come from traditional police work, not passive surveillance of every message. What changed is the ask. Agencies now want mandated backdoors: technical access built into encrypted systems for lawful requests. The paper argues this repeats the same conceptual error. A backdoor is not a feature you can limit to good actors. It is a vulnerability every attacker will find. The fifteen year gap is useful. It shows what actually happened when encryption went mainstream, not what lobby groups predicted. Criminals did not become invisible. They became harder to surveil in bulk, which is the point of encryption in the first place. The debate framing is still wrong. It treats encryption as a problem to solve, not a baseline security requirement. Every argument for weakening it ignores that hospitals, banks, and infrastructure also use E2EE. A backdoor for one investigation is a hole in every system.

---

Source: Encryption and Globalization 15 Years Later: E2EE and the “Going Dark” Debate