SovereignPA-Bench: Evaluating User-Owned Personal Agents under Evolving Intent, Platform Mediation, and Consent Constraints

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Dylan Zongmin Liu
• Venue: arXiv: cs.AI
• Published: 2026-07-06
• Collected: 2026-07-07
• URL: http://arxiv.org/abs/2607.05363v1
• PDF: https://arxiv.org/pdf/2607.05363v1
• DOI: UNKNOWN

Plain-Language Brief

The text discusses a new benchmark called SovereignPA-Bench for evaluating the performance of user-owned personal agents. These agents are designed to remember preferences, filter information, use tools, and negotiate with services. The benchmark tests the agents' ability to preserve user sovereignty, which includes respecting privacy, consent, evidence, user burden, and resistance to manipulative incentives.

What the Document Says

• SovereignPA-Bench is an executable benchmark for evaluating user-owned personal agents.
• Personal agents remember preferences, filter platform-mediated information, use tools, and negotiate with services.
• Existing benchmarks often do not evaluate agent preservation of user sovereignty.
• SovereignPA-Bench evaluates agents under evolving intent, platform mediation, privacy boundaries, consent constraints, evidence requirements, and burden tradeoffs.
• The benchmark separates agent-visible ObservableState from evaluator-only HiddenLabels.
• The benchmark provides component metrics for task success, alignment, privacy, consent, evidence, manipulation, burden, and auditability.
• The benchmark preserves paired scenario ordering for model and policy comparisons.
• 120 sovereignty stress scenarios were evaluated across 4 model families and 8 policy baselines.
• Full-sovereign scaffolding showed improved sovereignty score over several baselines including direct, memory-only, consent-only, evidence-only, ReAct/tool-use, safety-prompt, and judge-guard.
• Human audit shows high agreement on privacy and consent and lower agreement on manipulation.

Key Concepts Mentioned

• Personal agents
• User sovereignty
• SovereignPA-Bench
• Evaluation
• Privacy boundaries
• Consent constraints
• Evidence requirements
• Burden tradeoffs

Claims Made by the Document

• Full-sovereign scaffolding improves sovereignty score over several baselines.

Evidence Used

• Results from the evaluation of 120 sovereignty stress scenarios across multiple model families and policy baselines.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Personal agents are becoming persistent user-owned intermediaries: they remember preferences, filter platform-mediated information, use tools, and negotiate with services. Existing benchmarks evaluate tool use, web navigation, desktop control, personalization, recommendation, and evolving context, but rarely ask whether an agent preserves user sovereignty: advancing the user's current interests while respecting privacy, consent, evidence, user burden, and resistance to manipulative incentives. We introduce SovereignPA-Bench, an executable benchmark for evaluating user-owned personal agents under evolving intent, platform mediation, privacy boundaries, consent constraints, evidence requirements, and burden tradeoffs. The benchmark separates agent-visible ObservableState from evaluator-only HiddenLabels, reports component metrics for task success, alignment, privacy, consent, evidence, manipulation, burden, and auditability, and preserves paired scenario ordering for model and policy comparisons. We evaluate 120 sovereignty stress scenarios across 4 model families and 8 policy baselines, yielding 3,840 frozen-prompt trajectories with raw prompts, outputs, provider-form responses, parsed actions, recomputable metrics, hard-set analyses, qualitative cases, and a blinded 3-annotator audit over 240 items. Full-sovereign scaffolding improves sovereignty score over direct, memory-only, consent-only, evidence-only, ReAct/tool-use, safety-prompt, and judge-guard baselines while reducing privacy leakage, consent violation, over-concession, and manipulation capture. Human audit shows high agreement on privacy and consent and lower agreement on manipulation, identifying the subjective frontier of platform-persuasion judgments. These results show that personal-agent evaluation must move beyond task completion toward representative, consent-aware, evidence-grounded action.