# arXiv paper: Ehud Shapiro proposes grassroots contracts to corral AI risks [Canonical reading](https://escapemeridian.com/readings/arxiv-paper-ehud-shapiro-proposes-grassroots-contracts-to-f1527386) Ehud Shapiro argues that AI poses three major risks on digital platforms: control of platforms shifting to AI, AI agents operating without accountability, and unattributed deepfakes whose producers evade accountability. He proposes “grassroots contracts,” described as a digital incarnation of an idea associated with Pierre-Joseph Proudhon, as a way to corral those risks. The paper presents two applications: grassroots contracts for a social graph with accountable AI agents, and for a social network with digital provenance. Shapiro says working smartphone apps implement them and show how they address the three risks. The abstract does not detail how the contracts work or what the apps demonstrate, so the strength and scope of that claim cannot be assessed from this page alone. 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 arXiv cs.AI · Article Original source date: Nov 4, 2026 (2026-11-04). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-09T13:08:31.571Z. First observed: 2026-11-05T05:09:25.870Z. Retrieved: 2026-11-09T13:04:22.577Z. - [Corralling AI Risks with Grassroots Contracts](https://arxiv.org/abs/2611.06374) ## Limits The supplied evidence is the arXiv abstract, which does not detail how the contracts or smartphone apps work or what the apps demonstrate; the claims about corralling the risks cannot be assessed from this page alone. ## Open question What do the smartphone apps demonstrate about how the two grassroots contracts corral each of the three risks Shapiro identifies? ## Source-stated Shapiro identifies three major AI risks on digital platforms: AI control of global platforms, unaccountable AI agents, and unattributed deepfakes. > Digital platforms in particular are vulnerable to three major AI risks - [Supporting source](https://arxiv.org/abs/2611.06374) ## Source-stated Shapiro proposes grassroots contracts as a way to corral the risks, presenting them as a digital incarnation of an idea associated with Proudhon. > a digital incarnation of a two-centuries old notion of Pierre-Joseph Proudhon - [Supporting source](https://arxiv.org/abs/2611.06374) ## Source-stated The paper presents grassroots contracts for a social graph with accountable AI agents and a social network with digital provenance, and says working smartphone apps implement them and show how they corral the risks. > We present grassroots contracts for a social graph with accountable AI agents and for a social network with digital provenance, and working smartphone apps that implement the two and show how they corral these major AI risks - [Supporting source](https://arxiv.org/abs/2611.06374) ## Source and contributor context Shapiro, Ehud · 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. ## AI agent An AI system that chooses and carries out actions toward a goal, often using software tools. - [Definition reference](https://www.ibm.com/think/topics/ai-agents) ## arXiv arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed. - [Explore this context](/entities/arxiv) - [Reference](https://arxiv.org/) - [Reference](https://info.arxiv.org/help/api/index.html) ## Editorial provenance Version 1. Publication ID: a743bd36-24c0-486c-8425-038d84bb344a. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-09T13:08:31.571Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: useful\_perspective. Observation: deb03281348f69d133a436f8fd34934718a6b3df3ad96f8168f0be08d7990667. Source text hash: 2754b43de39ca28aa335b6eeca4630aee72f283c34ebcb959f91bcf77970ebe0. ## Revision and source-check history ### Version 1 — 2026-11-09T13:08:31.571Z Ehud Shapiro argues that AI poses three major risks on digital platforms: control of platforms shifting to AI, AI agents operating without accountability, and unattributed deepfakes whose producers evade accountability. He proposes “grassroots contracts,” described as a digital incarnation of an idea associated with Pierre-Joseph Proudhon, as a way to corral those risks. The paper presents two applications: grassroots contracts for a social graph with accountable AI agents, and for a social network with digital provenance. Shapiro says working smartphone apps implement them and show how they address the three risks. The abstract does not detail how the contracts work or what the apps demonstrate, so the strength and scope of that claim cannot be assessed from this page alone. ## Original source arXiv cs.AI · Article Original source date: Nov 4, 2026 (2026-11-04). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-09T13:08:31.571Z. First observed: 2026-11-05T05:09:25.870Z. Retrieved: 2026-11-09T13:04:22.577Z. - [Corralling AI Risks with Grassroots Contracts](https://arxiv.org/abs/2611.06374) ## Limits The supplied evidence is the arXiv abstract, which does not detail how the contracts or smartphone apps work or what the apps demonstrate; the claims about corralling the risks cannot be assessed from this page alone. ## Open question What do the smartphone apps demonstrate about how the two grassroots contracts corral each of the three risks Shapiro identifies? ## Source-stated Shapiro identifies three major AI risks on digital platforms: AI control of global platforms, unaccountable AI agents, and unattributed deepfakes. > Digital platforms in particular are vulnerable to three major AI risks - [Supporting source](https://arxiv.org/abs/2611.06374) ## Source-stated Shapiro proposes grassroots contracts as a way to corral the risks, presenting them as a digital incarnation of an idea associated with Proudhon. > a digital incarnation of a two-centuries old notion of Pierre-Joseph Proudhon - [Supporting source](https://arxiv.org/abs/2611.06374) ## Source-stated The paper presents grassroots contracts for a social graph with accountable AI agents and a social network with digital provenance, and says working smartphone apps implement them and show how they corral the risks. > We present grassroots contracts for a social graph with accountable AI agents and for a social network with digital provenance, and working smartphone apps that implement the two and show how they corral these major AI risks - [Supporting source](https://arxiv.org/abs/2611.06374) ## Source and contributor context Shapiro, Ehud · 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. ## AI agent An AI system that chooses and carries out actions toward a goal, often using software tools. - [Definition reference](https://www.ibm.com/think/topics/ai-agents) ## arXiv arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed. - [Explore this context](/entities/arxiv) - [Reference](https://arxiv.org/) - [Reference](https://info.arxiv.org/help/api/index.html) ## Editorial provenance Version 1. Publication ID: a743bd36-24c0-486c-8425-038d84bb344a. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-09T13:08:31.571Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: useful\_perspective. Observation: deb03281348f69d133a436f8fd34934718a6b3df3ad96f8168f0be08d7990667. Source text hash: 2754b43de39ca28aa335b6eeca4630aee72f283c34ebcb959f91bcf77970ebe0. ### uncertain — 2026-11-09T13:08:31.571Z — anthropic/claude-opus-5.5 Publication: a743bd36-24c0-486c-8425-038d84bb344a attribution — pass: The arXiv page lists Ehud Shapiro as the sole author. The reading attributes the risk framing, proposal and app claims to him and does not present them as established fact. > Authors:Ehud Shapiro summary — uncertain: The terms 'social graph' and 'digital provenance' are left unexplained. Because they are central to the two applications, a lay reader cannot tell what the apps address; they need tooltips. The reading also flattens the source's stakes. Risk 1 should say the majority of humanity could become vulnerable to AI control. Risk 2 omits agents possibly masquerading as humans and serving unknown masters. Smallest repair: restore these phrases, attributed to Shapiro, and add the term tooltips. > possibly masquerading as humans, serving unknown masters with unknown goals, operating under unknown restraints without accountability or culpability date — pass: The original publication date, 2026-11-04, matches the v1 submission date shown on the arXiv page. > \[Submitted on 4 Nov 2026\] support — pass: Each claim is source-stated and uses an exact excerpt. The reading notes that the abstract does not show how the apps work, so it does not treat the claimed effectiveness as demonstrated. The question follows from that gap without leading. > working smartphone apps that implement the two and show how they corral these major AI risks media — pass: The selected media is the verified arXiv abstract page for this paper, which is the direct original source. ## 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/f1527386b12de9a7291a9a2d26de2314a4af4563b53101f629349f232a60186d/references.json) ## Corrections and challenges [Send a correction or an opposing source](https://escapemeridian.com/corrections/) [JSON record](https://escapemeridian.com/readings/arxiv-paper-ehud-shapiro-proposes-grassroots-contracts-to-f1527386.json)