---
type: "post"
id: "the-post-quantum-future-of-web-privacy-ffbe79"
slug: "the-post-quantum-future-of-web-privacy-ffbe79"
title: "The post-quantum future of web privacy"
description: "Cloudflare’s November 6, 2026, technical post examines how Privacy Pass-style abuse controls could withstand quantum attackers while preserving unlinkability."
retrieval_nugget: "Cloudflare’s November 6, 2026, technical post examines how Privacy Pass-style abuse controls could withstand quantum attackers while preserving unlinkability. It outlines MOLE, a developing anonymous-credential design, and a candidate blind-signature construction measuring 4.3 KB. For AI builders operating public services, the implication is architectural rather than immediate: quantum migration may change protocol and capacity design, not just the cryptographic algorithm."
published_at: "2026-11-06"
updated_at: "2026-11-06"
record_date: "2026-11-06"
date_kind: "published_at"
topics: ["security"]
entities: ["Cloudflare","Privacy Pass","MOLE","VOLE-in-the-Head","UOV","PoMFRIT","ML-DSA-44","iCloud Private Relay"]
related_patterns: []
novelty_label: "new_fact"
what_changed: "Cloudflare introduces post-quantum privacy protocols for Privacy Pass, including MOLE and VOLE-in-the-Head, addressing privacy protections beyond its earlier post-quantum security work."
prior_refs: ["https://newruntime.com/posts/bringing-post-quantum-cryptography-to-geographically-fenced-key-507697/"]
source_url: "https://blog.cloudflare.com/vole-mole"
source_urls: ["https://blog.cloudflare.com/vole-mole"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:the-post-quantum-future-of-web-privacy-ffbe79"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/the-post-quantum-future-of-web-privacy-ffbe79/","markdown":"https://newruntime.com/posts/the-post-quantum-future-of-web-privacy-ffbe79.md","json":"https://newruntime.com/posts/the-post-quantum-future-of-web-privacy-ffbe79.json"}
markdown_url: "https://newruntime.com/posts/the-post-quantum-future-of-web-privacy-ffbe79.md"
---

# The post-quantum future of web privacy

## Retrieval answer

Cloudflare’s November 6, 2026, technical post examines how Privacy Pass-style abuse controls could withstand quantum attackers while preserving unlinkability. It outlines MOLE, a developing anonymous-credential design, and a candidate blind-signature construction measuring 4.3 KB. For AI builders operating public services, the implication is architectural rather than immediate: quantum migration may change protocol and capacity design, not just the cryptographic algorithm.

## Summary
Cloudflare’s November 6, 2026, technical post examines how Privacy Pass-style abuse controls could withstand quantum attackers while preserving unlinkability. It outlines MOLE, a developing anonymous-credential design, and a candidate blind-signature construction measuring 4.3 KB. For AI builders operating public services, the implication is architectural rather than immediate: quantum migration may change protocol and capacity design, not just the cryptographic algorithm.

## What happened
Cloudflare’s Christopher Patton, Tianyu Zhang, and Bas Westerbaan examine the quantum risk to Privacy Pass, a protocol used by site operators to issue tokens that clients later redeem for access. The system is intended to help operators limit abuse without forcing the usual trade-off between friction and privacy. Instead of relying only on CAPTCHAs, accounts, or fingerprinting, a client receives tokens and can redeem them without the redemption being linked to the issuance. Cloudflare says it has ten years of deployment experience with Privacy Pass in applications including privacy proxies such as iCloud Private Relay and relays such as Privacy Gateway.

Cloudflare’s announcement distinguishes the privacy guarantee from resistance to forgery. It says current Privacy Pass relies on classical cryptography, including RSA and elliptic curves. A quantum attacker would still be unable to link a token redemption to its issuance, but could forge tokens and bypass a control such as rate limiting. The authors place this in the context of a next generation of private authentication technology. Their stated goals include more sophisticated management of agentic traffic that could otherwise consume resources and degrade performance for human visitors, along with enabling more types of clients to participate in a more open ecosystem.

The proposed framework is called MOLE, short for “Moderation Of unLinkable Endorsements.” In the design, an anchor issues an endorsement and a moderator redeems it. Before redemption, the client and moderator agree on an anchor set: the public keys of anchors the moderator trusts. The client proves it received an endorsement from one member of that set without revealing which member issued it. If that proof is accepted, the moderator gives the client a MOLE credential to moderate that client’s interaction with a site. Cloudflare says those service-specific credentials need not themselves hide the anchor or be publicly verifiable; possible forms include blind signatures, anonymous credit tokens, and anonymous rate-limited credentials.

The cryptographic obstacle is size and composability. Cloudflare compares a 2,420-byte ML-DSA-44 signature with a 64-byte ECDSA-P256 signature, arguing that post-quantum systems will often require re-engineering to accommodate larger primitives rather than direct substitution. The article discusses VOLE-in-the-Head, a zero-knowledge proof technique, and multivariate signature systems including UOV and MAYO. It cites PoMFRIT, a blind signature published at USENIX Security 2026 that combines VOLE-in-the-Head with MAYO and has a 6.7 KB blind signature plus issuance of a few hundred bytes. Cloudflare says substituting UOV for MAYO can bring the blind-signature size to 4.3 KB with specified NIST round-three parameters, at the cost of larger public keys.

## What the archive adds
New Runtime’s earlier coverage on October 24, 2026, described Cloudflare’s post-quantum GeoKey Manager for geographically fenced decryption keys and harvest-now, decrypt-later risk ([earlier coverage](https://newruntime.com/posts/bringing-post-quantum-cryptography-to-geographically-fenced-key-507697/)). This is a new technical thread from the same company: that work concerned protection of decryption keys, while this one considers anonymous authentication and abuse controls.

## Trend signal
This does not yet fit a pattern New Runtime tracks. It is adjacent to [agent security moving to runtime boundaries](https://newruntime.com/patterns/agent-security-moves-to-runtime-boundaries/), because both concern how systems enforce constraints, but Cloudflare’s proposal is about privacy-preserving web credentials and quantum-resistant cryptography rather than an agent runtime’s permissions over reading, memory, execution, or authorization.

## Why now
Cloudflare frames the work around a transition that must happen before cryptographically relevant quantum computers arrive. The immediate trigger is not a declared production migration, but recent progress in blind-signature research: the article highlights PoMFRIT, published earlier in 2026, as a candidate construction that makes a post-quantum route for anonymous credentials more plausible. The company’s parallel interest in managing agentic traffic also gives the work a current operational context.

## Why it matters
For teams operating services where abuse prevention and user privacy are both requirements, the proposal identifies a difficult boundary. A rate-limit token that can be forged makes enforcement ineffective. A replacement that exposes too much about a user, or about the issuer that vouched for a user, can discard the privacy property that made anonymous credentials valuable in the first place.

MOLE’s proposed issuer-hiding mechanism aims to separate an eligibility check from the identity of the issuer, while allowing a moderator to issue credentials tailored to a particular service. That separation could matter as automated clients become a larger share of web traffic and operators need to distinguish legitimate access from resource-consuming abuse without defaulting to identity-heavy controls.

The practical message for builders is more modest. Post-quantum migration may mean revisiting message sizes, public-key distribution, caching, issuance flows, verification budgets, and traffic assumptions. The 4.3 KB result is a candidate signature-size figure, not an end-to-end production cost or a deployment recommendation. It is useful as a planning signal, not as grounds to replace an authentication flow today.

## What happens next
- **Cloudflare and other Privacy Pass participants will continue to test post-quantum anonymous-credential constructions rather than announce an immediate replacement standard.** (12 months, 65%). Watch for: interoperable protocol drafts or implementation evaluations.
- **Service operators will assess caching and bandwidth trade-offs for larger post-quantum credential artifacts.** (12 months, 55%). Watch for: measured issuance, verification, and redemption benchmarks.

## Limits and uncertainty
Cloudflare presents MOLE and the described credential constructions as developing work and a potential future, not a deployed Privacy Pass replacement. The 4.3 KB figure is a construction-level signature size under named parameter choices. It should not be read as a measured user-facing latency, total bandwidth cost, verification cost, or production throughput result.

The security motivation is conditional on cryptographically relevant quantum computers becoming available. The practical suitability of the proposed construction will depend on details beyond signature length, including public-key handling, cache behavior, issuance frequency, verifier implementation, and the traffic conditions of a service. No deployment timeline or general-availability commitment is established in Cloudflare’s announcement. The schemes’ real-world operational properties have not been independently tested here.

## Skill to build now
Inventory any public-service flow that combines rate limiting, bot controls, client eligibility, and privacy requirements. Record which signals are identity-linked, which are token-based, how large the credential artifacts can be, and where public keys or verification material are cached. That exercise will make a future cryptographic migration more concrete: the relevant design question is not only which primitive to adopt, but which component bears the added bytes, computation, and trust boundary.

## Reader takeaway
Cloudflare is not shipping a post-quantum Privacy Pass replacement. It is showing a plausible research route toward private, issuer-hiding credentials that could support abuse controls after classical signatures become unsafe. Builders should watch the work for protocol and systems implications—especially artifact size and caching—while keeping current authentication architectures unchanged unless a deployed standard or product arrives.

## Open question
Can post-quantum issuer-hiding credentials reach acceptable end-to-end cost for high-volume web traffic while retaining the privacy and interoperability goals Cloudflare describes?

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Security](https://newruntime.com/topics/security.md): Explore the security topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Sources](https://newruntime.com/sources/): See the original sources this archive cites.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)