# Marginal Revolution shares a worst-case scenario for AI hacking risks [Canonical reading](https://escapemeridian.com/readings/marginal-revolution-shares-a-worst-case-scenario-for-ai-hacking-e33ab201) Marginal Revolution shares an excerpt from a Free Press column that presents a worst-case scenario, not a prediction, about cryptographic protections becoming vulnerable. The excerpt argues that protections would likely fail unevenly: public key encryption might become vulnerable first, while other safeguards and existing backup systems could limit the damage. It says the shift to more secure communication could be costly, with smaller businesses facing more serious transition problems and individuals potentially having to visit banks in person for major transactions. The author also suggests that advances in mathematics might provide warning before a large-scale attack, giving institutions time to adjust. A period of uncertainty could still prompt financial shifts and attempts to capture confidential material for later decryption. The excerpt does not establish that such a breakthrough or attack will occur. Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims. ## Original source Marginal Revolution · Article Original source date: Oct 28, 2026 (2026-10-28T04:38:22+00:00). Date basis: Date stated by the publisher. Added to Escape Meridian: 2026-11-05T07:31:57.724Z. First observed: 2026-10-28T04:48:56.744Z. Retrieved: 2026-10-28T05:00:32.643Z. - [AI hacking risks - Marginal REVOLUTION](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Limits The source is an excerpt from a Free Press column, so the full column's context and supporting analysis are not included. The excerpt presents hypothetical scenarios; it does not establish that cryptographic protections will become vulnerable or that a large-scale attack will occur. ## Open question What observable mathematical or cryptographic developments would give institutions enough warning to adjust before a large-scale attack? ## Source-stated The excerpt explicitly frames the scenario as a worst case rather than a prediction. > noting this is a worst-case scenario not a prediction: - [Supporting source](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Source-stated The author says public key encryption might become vulnerable before other protections, and that a shift toward more secure communication could be costly. > The more likely scenario, assuming this comes to pass at all, is that public key encryption would become vulnerable first - [Supporting source](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Source-stated The author argues that intermediate mathematical advances could signal vulnerability and provide time to adjust. > intermediate advances may signal that cryptographic protections are vulnerable. We thus would have some time to adjust… - [Supporting source](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Source-stated The author describes possible financial shifts and attempts to capture confidential material for later decryption if concerns about encryption grew. > Some hackers would invest more in the activity, including hacks without advanced AI, under the philosophy of “harvest now, decrypt later,” - [Supporting source](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Source and contributor context Tyler Cowen · Article author These are credits stated by the source; they do not establish a profile match or authorship of every passage. ## Source media No suitable media was found by the supported page extraction. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Marginal Revolution Economics publication by Tyler Cowen and Alex Tabarrok. - [Explore this context](/entities/marginal-revolution) - [Reference](https://marginalrevolution.com/) ## Tyler Cowen Economics, institutions and unexpected cross-domain questions; useful source discovery. - [Explore this context](/entities/tyler-cowen) - [Reference](https://www.dwarkesh.com/p/tyler-cowen-4) - [Reference](https://www.dwarkesh.com/p/tyler-cowen-3) - [Reference](https://www.dwarkesh.com/p/tyler-cowen-2) - [Reference](https://marginalrevolution.com/marginalrevolution/2026/10/saturday-assorted-links-583.html?utm_source=rss&utm_medium=rss&utm_campaign=saturday-assorted-links-583) - [Reference](https://tylercowen.com/) ## Editorial provenance Version 1. Publication ID: fead3834-2723-4787-835e-89bfcc24c61c. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-05T07:31:57.724Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: useful\_perspective. Observation: 16c3b76bdd5d58bd63bd0b6ee30a86b9f7ca105cf2d24ef7a446e0d2242f3f44. Source text hash: 5aa64b050212bf5acc8e9c8eeefdeea924c554a9a60bbe7aecf50b8636dad9d0. ## Revision and source-check history ### Version 1 — 2026-11-05T07:31:57.724Z Marginal Revolution shares an excerpt from a Free Press column that presents a worst-case scenario, not a prediction, about cryptographic protections becoming vulnerable. The excerpt argues that protections would likely fail unevenly: public key encryption might become vulnerable first, while other safeguards and existing backup systems could limit the damage. It says the shift to more secure communication could be costly, with smaller businesses facing more serious transition problems and individuals potentially having to visit banks in person for major transactions. The author also suggests that advances in mathematics might provide warning before a large-scale attack, giving institutions time to adjust. A period of uncertainty could still prompt financial shifts and attempts to capture confidential material for later decryption. The excerpt does not establish that such a breakthrough or attack will occur. ## Original source Marginal Revolution · Article Original source date: Oct 28, 2026 (2026-10-28T04:38:22+00:00). Date basis: Date stated by the publisher. Added to Escape Meridian: 2026-11-05T07:31:57.724Z. First observed: 2026-10-28T04:48:56.744Z. Retrieved: 2026-10-28T05:00:32.643Z. - [AI hacking risks - Marginal REVOLUTION](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Limits The source is an excerpt from a Free Press column, so the full column's context and supporting analysis are not included. The excerpt presents hypothetical scenarios; it does not establish that cryptographic protections will become vulnerable or that a large-scale attack will occur. ## Open question What observable mathematical or cryptographic developments would give institutions enough warning to adjust before a large-scale attack? ## Source-stated The excerpt explicitly frames the scenario as a worst case rather than a prediction. > noting this is a worst-case scenario not a prediction: - [Supporting source](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Source-stated The author says public key encryption might become vulnerable before other protections, and that a shift toward more secure communication could be costly. > The more likely scenario, assuming this comes to pass at all, is that public key encryption would become vulnerable first - [Supporting source](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Source-stated The author argues that intermediate mathematical advances could signal vulnerability and provide time to adjust. > intermediate advances may signal that cryptographic protections are vulnerable. We thus would have some time to adjust… - [Supporting source](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Source-stated The author describes possible financial shifts and attempts to capture confidential material for later decryption if concerns about encryption grew. > Some hackers would invest more in the activity, including hacks without advanced AI, under the philosophy of “harvest now, decrypt later,” - [Supporting source](https://marginalrevolution.com/marginalrevolution/2026/10/ai-hacking-risks.html) ## Source and contributor context Tyler Cowen · Article author These are credits stated by the source; they do not establish a profile match or authorship of every passage. ## Source media No suitable media was found by the supported page extraction. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Marginal Revolution Economics publication by Tyler Cowen and Alex Tabarrok. - [Explore this context](/entities/marginal-revolution) - [Reference](https://marginalrevolution.com/) ## Tyler Cowen Economics, institutions and unexpected cross-domain questions; useful source discovery. - [Explore this context](/entities/tyler-cowen) - [Reference](https://www.dwarkesh.com/p/tyler-cowen-4) - [Reference](https://www.dwarkesh.com/p/tyler-cowen-3) - [Reference](https://www.dwarkesh.com/p/tyler-cowen-2) - [Reference](https://marginalrevolution.com/marginalrevolution/2026/10/saturday-assorted-links-583.html?utm_source=rss&utm_medium=rss&utm_campaign=saturday-assorted-links-583) - [Reference](https://tylercowen.com/) ## Editorial provenance Version 1. Publication ID: fead3834-2723-4787-835e-89bfcc24c61c. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-05T07:31:57.724Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: useful\_perspective. Observation: 16c3b76bdd5d58bd63bd0b6ee30a86b9f7ca105cf2d24ef7a446e0d2242f3f44. Source text hash: 5aa64b050212bf5acc8e9c8eeefdeea924c554a9a60bbe7aecf50b8636dad9d0. ### uncertain — 2026-11-05T07:31:57.724Z — anthropic/claude-opus-5.5 Publication: fead3834-2723-4787-835e-89bfcc24c61c attribution — pass: The draft attributes the excerpt to a Free Press column shared on Marginal Revolution and refers to 'the author' in a first-person post. The author is not named in the text beyond a 'thefp.com/tyler' link, so attributing to the publisher and author is acceptable and does not misattribute claims. > That is the topic of my latest Free Press column (now there are gift links!) summary — uncertain: The summary never explains the AI link that gives the piece its significance: the threat is AI-driven math breakthroughs, illustrated by OpenAI proving a quasi-Riemann hypothesis. It also leaves 'public key encryption' and 'harvest now, decrypt later' unexplained, and drops the author's Covid analogy and blunt 'it would not be good,' flattening his voice. Smallest repair: one sentence that AI-found math techniques could break encryption, citing the OpenAI/Tao example, plus a plain gloss of public key cryptography. > OpenAI has proven a quasi-Riemann hypothesis, and Terry Tao told me recently that this may open the door to solving the Riemann hypothesis proper. date — pass: originalPublished 2026-10-28T04:38:22Z comes from the publisher page and matches the feed entry. Nothing indicates an earlier first publication of this post. support — pass: The claims match the source: worst case, not a prediction; public key encryption vulnerable first; costly transition; smaller businesses face harder transitions; in-person bank visits; intermediate advances giving time to adjust; financial shifts and harvest-now-decrypt-later. The question follows from the source's warning argument without assuming an answer. Limitations properly preserve the hypothetical status. > The transition would be very costly, but the end result would not be total insecurity. media — pass: The selected media is the verified direct link to the original Marginal Revolution post, which matches the source URL. No better source media was found. ## Recorded references Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter. No public references recorded here yet. [Reference data (JSON)](/readings/e33ab201dc691c30454196a7ad81eeee4578cf33e6026a675e676d7cb0e221bc/references.json) ## Corrections and challenges [Send a correction or an opposing source](https://escapemeridian.com/corrections/) [JSON record](https://escapemeridian.com/readings/marginal-revolution-shares-a-worst-case-scenario-for-ai-hacking-e33ab201.json)