Is it ethical to collect data today to decrypt tomorrow the morality of patient spying

[TrueRoss42]

TL;DR

• Harvest now, decrypt later describes collecting encrypted data today in the hope of reading it once quantum computers or other advances make decryption possible
• Intelligence agencies have long intercepted communications they could not immediately read, and Edward Snowden's 2013 disclosures showed the scale of modern bulk collection
• Ethical frameworks for intelligence, such as the principles proposed by former GCHQ director Sir David Omand, emphasise just cause, proportionality, right authority and necessity
• In 2021, the European Court of Human Rights ruled in Big Brother Watch v United Kingdom that bulk interception is not inherently unlawful but requires strong safeguards, which the UK regime at the time lacked
• Reasonable people disagree: some see patient collection as prudent national security, others as a mass privacy violation waiting to happen

The patient spy

Spying is often imagined as dramatic: secret agents, stolen documents and daring escapes. Much modern intelligence work is quieter. It involves collecting vast amounts of communications, storing them and analysing them over time. Some of that data may be encrypted and unreadable today, yet still worth keeping in case it can be read later

Security experts call this harvest now, decrypt later. The idea is that a future quantum computer, or some other breakthrough, might unlock data encrypted with today's public key methods. An adversary willing to store data for years could eventually read diplomatic cables, business secrets or personal messages. This is one of the main reasons governments are urging rapid migration to post quantum cryptography

This thread asks a moral question rather than a technical one. Is it ethical to collect data you cannot read in the hope of reading it later? Does the answer change depending on who collects it, whose data it is and what they plan to do? I hope members will share their perspectives, including disagreements

A long tradition

Intercepting communications that cannot yet be read is not new. During the Cold War, the US Venona project collected Soviet messages for decades, gradually decrypting portions as analysts found weaknesses. Codebreakers throughout history have stored intercepts, hoping that new techniques or captured keys would eventually unlock them. Patience has always been part of signals intelligence

What has changed is scale. The internet carries an enormous volume of communications, and storage has become cheap. In 2013, documents disclosed by former NSA contractor Edward Snowden revealed programmes of bulk collection by US and allied agencies, sparking global debate about surveillance. While the specifics of what is stored and for how long remain secret, the potential for long term retention is clear

Quantum computing adds a twist. Encryption that once seemed permanently secure might become breakable within the lifetime of the data. That turns stored ciphertext from an archive of noise into a potential future treasure. It also changes the ethics, because collection today becomes a bet on access tomorrow

Private companies collect data too. Technology firms store vast amounts of user communications and metadata, and some of that data is encrypted with methods that could become vulnerable. If such stores were stolen, harvest now, decrypt later could apply to criminals as well as states. The ethics of long term retention therefore extend beyond intelligence agencies

Ethical frameworks for intelligence

Philosophers and practitioners have developed frameworks for judging intelligence activities. Sir David Omand, a former director of the UK's signals intelligence agency GCHQ, has proposed principles adapted from just war thinking, discussed in his writing including the 2018 book Principled Spying co-authored with Mark Phythian. They include a sufficient and sustainable cause, integrity of motive, proportionality, right authority, a reasonable prospect of success and necessity, meaning that less intrusive methods would not suffice. These principles offer a useful lens for harvest now, decrypt later

Applying them raises hard questions. Is collecting vast amounts of encrypted traffic proportionate when most of it concerns ordinary people? Is there right authority, meaning proper legal authorisation and oversight? Is there a reasonable prospect of success, given that decryption may never become possible for some data? Thinking through these questions reveals why the practice is contested

Other frameworks emphasise rights. International human rights law, including Article 17 of the International Covenant on Civil and Political Rights, protects against arbitrary or unlawful interference with privacy and correspondence. Any surveillance must have a legal basis and be necessary and proportionate. Collecting data with the intention of reading it later sits uneasily with these principles unless strong safeguards apply

Consent is notably absent. People whose encrypted messages are collected have no idea it is happening and no chance to object. In most areas of life, collecting someone's private information without consent requires strong justification. Intelligence work is an exception that societies accept only under strict conditions

What the courts have said

Courts have grappled with bulk surveillance. In May 2021, the Grand Chamber of the European Court of Human Rights ruled in Big Brother Watch and Others v the United Kingdom. It held that bulk interception is not inherently incompatible with the European Convention on Human Rights, given the threats states face, but that it must be subject to end to end safeguards. The court found that the UK's regime at the time lacked sufficient safeguards and therefore violated the right to privacy

The judgment reflects a balance. It accepts that states may need broad collection capabilities, while insisting on independent authorisation, clear rules on selection and retention, and effective oversight. The UK had already replaced much of the old framework with the Investigatory Powers Act 2016, which created the Investigatory Powers Commissioner. How such safeguards apply to long term retention of encrypted data is a question few laws address explicitly

International law on espionage between states is notably ambiguous. Experts writing the Tallinn Manual 2.0, published in 2017, concluded that peacetime espionage is not in itself a violation of international law, although methods used may breach other rules. States therefore tend to condemn being spied on while reserving the right to spy. Harvest now, decrypt later fits into this long standing grey zone

Arguments that patient collection can be justified

Supporters argue that intelligence agencies have a duty to protect citizens from serious threats such as terrorism, espionage and military aggression. Adversaries' communications may contain information vital to national security, even years later. Collecting encrypted data that might one day be readable is a reasonable precaution, much like keeping evidence that may become useful. Failing to collect it could mean missing critical insights

They also point out that adversaries certainly do it. If one country refrains while others collect aggressively, it may be placed at a disadvantage. In a world without binding international agreements, unilateral restraint can be costly. Responsible collection under law and oversight is, on this view, better than leaving the field to less scrupulous actors

Some argue that targeted retention can be proportionate. Storing encrypted communications of specific foreign intelligence targets, under proper authorisation, differs from storing everyone's data indefinitely. Careful scoping and deletion rules could address many concerns. The ethics depend heavily on how the practice is carried out

Retention limits are a practical safeguard. Laws can require that collected data be deleted after a set period unless it is specifically needed. Such limits reduce the reservoir available for future decryption. Whether they would apply to data stored precisely because it might be decrypted later is an open question

Arguments that it crosses a line

Critics argue that bulk collection of encrypted data treats everyone as someone under suspicion. Most stored communications belong to ordinary people with no connection to threats. Retaining them in the hope of future decryption creates a vast reservoir of private information that could be misused, leaked or exploited by future governments with different values. Data collected for one purpose rarely stays confined to it

They also highlight the chilling effect. If people believe their private messages may eventually be read, they may self censor, avoid sensitive topics or withdraw from public debate. Journalists, activists, lawyers and minority communities may feel this most strongly. A healthy democracy depends on spaces for private thought and communication

There is a further concern about accountability over time. Oversight bodies may approve collection today, but decryption could occur decades later under different leaders and laws. Those who authorised collection may be long gone, and the people affected may never know. Long time horizons make meaningful accountability difficult

Trust in institutions shapes how people react. In countries where intelligence agencies are seen as accountable and law abiding, people may accept more collection. Where trust is low, even modest collection can provoke anger. Public trust is itself a resource that careless practices can destroy

The defender's dilemma

Harvest now, decrypt later also creates an ethical duty for defenders. If adversaries may be storing traffic, organisations holding sensitive data have a responsibility to protect it with post quantum encryption as soon as practical. Governments that warn about the practice while doing it themselves face an awkward tension. Many security agencies nonetheless lead efforts to help everyone migrate

For ordinary people, the practical response is the same as for other quantum risks: use updated software that adopts post quantum protection, minimise sensitive data sent electronically and support strong privacy laws. The ethical debate about collection continues, but individuals and organisations can reduce their exposure regardless. Protecting data well is a moral as well as technical choice

Questions to discuss

• Is collecting encrypted data you cannot read today a form of surveillance, or only potential surveillance
• Should intelligence agencies be allowed to store encrypted traffic indefinitely, for a fixed period, or not at all
• Do Omand's principles convince you, and which matters most
• Is it hypocritical for governments to warn about harvest now, decrypt later while doing it themselves
• Would knowing your messages might be stored change how you communicate

Closing thoughts

Patient spying sits at the crossroads of national security, privacy and time. History shows that intercepted secrets can be read decades later, and quantum computing could make that more common. Ethical frameworks emphasise proportionality, authority and necessity, while courts insist on safeguards. Where the line should fall is a question every democracy must answer, ideally in open debate

Please share your views in the replies, whether you see patient collection as prudent, troubling or somewhere in between. This is a topic on which thoughtful people really differ. Respectful discussion helps us all understand the trade offs better


Further reading

• David Omand and Mark Phythian, Principled Spying: The Ethics of Secret Intelligence, 2018
• European Court of Human Rights, Big Brother Watch and Others v the United Kingdom, Grand Chamber judgment, May 2021
• UK Investigatory Powers Act 2016
• Michael Schmitt (ed.), Tallinn Manual 2.0 on the International Law Applicable to Cyber Operations, 2017
• International Covenant on Civil and Political Rights, Article 17

[alwaysPatrick19]

Until now I had never thought of storing unreadable data as surveillance. It seemed harmless if nobody could read it

Your post changed my mind somewhat. If the intent is to read it later, the collection is the first step of surveillance. The fact that the lock opens later does not make it innocent. Intent matters a great deal in ethics. Law often recognises that too

Interesting how a technical idea leads straight into philosophy. That is why I enjoy this forum

I will be thinking about this for a while. Thanks for the challenge

[PrimeToby37]

Omand's principles make a helpful checklist. Proportionality feels like the key one to me

Storing a known spy's encrypted messages seems proportionate. Storing everyone's seems far harder to justify

Most people would accept the first and reject the second. The problem is that bulk systems blur that line. Selection happens after collection, not before. That is what makes people uneasy

Strong deletion rules could help. But trusting them requires faith in oversight

[DelPiero58]

Having served in the armed forces, I see why agencies collect widely. Threats are real, and missing a clue can cost lives

That said, oversight matters enormously. I would rather have strong independent checks than no collection at all. Courts and commissioners give the public some assurance. Without them, abuses become likely

Balance is everything in this area. Neither extreme works. Good oversight is how balance is kept