---
格式版本: 2
标题: "Inferential Capability Does Not Determine Legal Scope"
原文链接: "https://arxiv.org/abs/2608.10601"
发布日期: "2026-08-11"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Tue, 11 Aug 2026 07:39:53 UTC (33 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-12T11:30:27+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-12T11:26:18+08:00"
入库时间: "2026-08-12T03:30:27.869Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=AI&searchtype=all"
匹配关键词:
  - "AI"
相关厂家:
  []
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 5
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "论文讨论欧盟AI法案与GDPR的推理法律定义，与超节点/AI Rack/机柜级AI基础设施完全无关。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-12T11:30:47+08:00"
AI主题相关性: 0
AI来源权威性: 5
AI新颖性: 0
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 0
采集批次: "2026年8月12日2点04分20秒"
采集批次ID: "20260812-020420-289"
去重键: "https://arxiv.org/abs/2608.10601"
---

## Computer Science > Computers and Society

## Title:Inferential Capability Does Not Determine Legal Scope

Authors:[Nicola Fabiano](https://arxiv.org/search/cs?searchtype=author&query=Fabiano,+N)

[View PDF](https://arxiv.org/pdf/2608.10601) [HTML (experimental)](https://arxiv.org/html/2608.10601v1)

> Abstract:Two instruments of EU digital law place inference at their centre and mean different things by it. Article 3(1) of the AI Act uses the capability to infer constitutively: it is the central feature separating the regulated category from conventional software. The GDPR never defines inference, yet governs it protectively: the consequences follow from the processing of personal data and from what the inference says about, or does to, a person, whether or not the technology that produced it qualifies as an AI system.  
> The two perimeters are not concentric. Their non-coincidence remained invisible in single-shot systems; agentic architectures make it operationally acute. The thesis: inferential capability does not determine legal scope, and its absence does not create immunity.  
> The framework is two-level. Inference performs two legal functions, constitutive and protective; the protective function operates through three pathways - identificatory, attributive and decisional. Composition is not a fourth pathway but a cross-cutting architectural dimension which, with reach, persistence and reviewability, is what agentic architectures modify. Three concepts support it: the inferential threshold, the inferential reach and the inferential chain, mapped onto the chain of imputation.  
> Regulation (EU) 2026/1744 left the constitutive criterion untouched and inserted a provision contemplating outputs that influence the inputs of future operations, without supplying any rule of aggregation. The article proposes an interpretive rule, a compositional-effects test identifying the decision unit under Article 22 GDPR together with the allocation of the burden of establishing it, and documentation duties calibrated to inference chains.

| Subjects: | Computers and Society (cs.CY); Artificial Intelligence (cs.AI) |
| --- | --- |
| Cite as: | [arXiv:2608.10601](https://arxiv.org/abs/2608.10601) \[cs.CY\] |
|  | (or [arXiv:2608.10601v1](https://arxiv.org/abs/2608.10601v1) \[cs.CY\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.10601](https://doi.org/10.48550/arXiv.2608.10601) |

## Submission history

From: Nicola Fabiano \[[view email](https://arxiv.org/show-email/22c376e5/2608.10601)\]  
**\[v1\]** Tue, 11 Aug 2026 07:39:53 UTC (33 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.10601) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))