Meta published its post-quantum migration playbook five days ago. The post-quantum cryptographic tools the internet needs to survive the quantum computing era have been sitting in open-source repositories for years — free, production-ready, and waiting. The thing your security vendor is selling you is not the cryptography. It is the organizational work of actually deploying it.
The Commercial National Security Algorithm Suite 2.0, or CNSA 2.0, requires post-quantum cryptography across US federal systems by January 1, 2027, Meta's playbook notes. That is nine months away. Most large organizations are still inventorying which TLS certificates need replacing. A medium-to-large organization typically needs three years to complete the work, a Computing UK survey found. They have nine months.
The open-source community built the plumbing. Enterprise is still trying to figure out which pipe connects to which wall.
When Meta enabled hybrid post-quantum TLS for internal services, its engineers found that the added cryptographic signature bloats the initial handshake by roughly 1,184 bytes. Corporate firewalls, load balancers, and inspection appliances built between 2015 and 2019 often drop or mangle fragmented ClientHellos, Meta's engineering team documented. The problem compounds under real-world conditions: under 3% packet loss, the larger PQC handshake adds retransmission overhead that Meta measured as a 32% latency increase over the classical baseline. The fix is not a cryptographic question. It is a question of which appliances in which rack need firmware updates, and whose budget that falls under.
There is a second problem hiding inside the certificate chain. When AWS Certificate Manager opened hybrid certificate support in 2025, its engineers discovered that legacy validators silently fail on dual-signature chains. The chain parses. The second signature, the post-quantum one, is ignored. The system reports full protection. There is none. Meta documented the same failure mode. The organizations that think they are furthest along are in some cases the ones with the most fragile false confidence.
Cloudflare reached majority post-quantum protection for human-initiated traffic in October 2025, the company announced that month. It is the closest thing to a success story in this space, and it took years of engineering work by a company whose core product is internet infrastructure. The lesson from that work is not that quantum-safe security is hard to build. It is that it is hard to retrofit onto a live, heterogeneous global network where a Fortune 500 company's internal tooling might include appliances from six different procurement cycles, none of which its security team fully inventoried until prompted.
The vendor ecosystem has noticed. Around half of UK IT leaders say they expect to be completely or heavily reliant on vendors for their quantum-safe transition, Computing UK found. That is rational. The question is what vendors are actually selling. The cryptographic algorithms are free. The Open Quantum Safe project's liboqs is MIT-licensed and integrated with OpenSSL, the most widely deployed TLS library in the world. The engineering work to ship post-quantum cryptography at internet scale has been done. What remains is change management: the auditing, the certificate rotation, the appliance firmware updates, the testing cycles, the procurement conversations. Hybrid certificate issuance windows at major public certificate authorities are opening in Q3 2026, but availability at scale will lag into 2027, Meta documented.
That work is not trivial. Three years is the estimate for good reason. But it is a category of work that does not require cryptographic innovation. It requires project managers, procurement approvals, and internal documentation that most organizations are only now beginning to compile.
Craig Gidney published revised quantum computing estimates in June 2025 showing that RSA-2048 could be broken with fewer than one million qubits, Cloudflare reported at the time, compressing earlier estimates by roughly seven years. That change in the threat model makes the organizational problem more urgent, not less. The window for solving this is not defined by when the open-source tools were ready. It is defined by when adversaries can realistically exfiltrate today's encrypted data and decrypt it later, a timeline that moved closer with Gidney's paper, not further.
The open-source plumbing shipped commoditized and free. The thing the federal government, the Fortune 500, and the managed security provider are actually negotiating is the gap between a working cryptographic library and a fully audited, deployed, and monitored TLS stack in a running enterprise network. That gap is real. It is large. And it is organizational, not cryptographic.