arXiv:2606.10794v1 Announce Type: new Abstract: As agentic applications increasingly route user tasks through official and third-party LLM APIs, provenance becomes an operational question: which model generated a given black-box response? We study Dynamic Black-Box LLM Provenance: identifying the source LLM from generations elicited by query-varying, non-predefined prompts rather than a fixed input set or benchmark suite. This setting is difficult because prompt semantics dominate the text, while model-specific authorship traces are weak and inconsistent at the surface level. We introduce READ
The proliferation of agentic applications and diverse LLM APIs makes identifying the source of generative AI outputs an immediate operational and security concern.
Understanding LLM provenance is critical for accountability, intellectual property, and combating misinformation in an increasingly AI-driven information environment.
The ability to reliably attribute LLM-generated content to its originating model shifts the landscape for content verification, legal liability, and AI security.
- · AI security firms
- · Content verification platforms
- · Regulatory bodies
- · Malicious actors using AI for disinformation
- · Platforms struggling with AI content moderation
- · Undisclosed AI model developers
Improved trust and accountability in AI-generated content through better attribution.
Development of new regulatory frameworks and industry standards for AI provenance and disclosure.
Potential for 'AI copy-detection' markets and a stronger defense against AI-driven intellectual property theft.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI