| ai.viXra.org > Artificial Intelligence > 2610 |
|
 |
| ai.viXra.org > Artificial Intelligence > 2610 |
|
|
[3] ai.viXra.org:2610.0049 [pdf] submitted on 2026-10-22 14:19:32
Authors: Davie Chen
Comments: 15 pages, 7 figures, 7 tables. Unreviewed preprint, version 2. CC BY 4.0. Zenodo DOI: 10.5281/zenodo.23556480.
Scientific assistants increasingly gather evidence through tools that aggregate multiple external sources. A record list alone does not reveal whether an empty result reflects a completed search, incomplete coverage, or total upstream failure. This paper develops a failure-transparency protocol that separates record identity, request coverage, and claim support. We combine three evidence streams: an audit of eighteen dated website examples containing twenty-four displayed records across eighteen source identifiers; an exhaustive analysis of 117 abstract source-state configurations; and a small live metadata-check episode. All displayed records carry a URL and source field, but the shortened examples do not expose a per-source execution envelope, illustrating why record provenance and coverage are distinct audit properties. In the controlled enumeration, record-only and HTTP-aware policies correctly classify 28 and 31 configurations respectively; a coverage-aware reference classifies all 117 by construction. These are contract checks, not language-model performance measurements. During three read-only Crossref requests, two returned HTTP 429 while one returned a retraction-notice record, providing a concrete case in which unavailable metadata must not be interpreted as a missing publication. Fifty-two offline connector checks also passed using previously recorded responses; this does not measure current upstream reliability. Illustrated reference, dataset, literature, and archive examples motivate a three-layer reporting contract and a practical answer-level evaluation rubric. The study offers a reproducible interface-audit method, not a new retrieval model or evidence of a commercial service's superiority.
Category: Artificial Intelligence
[2] ai.viXra.org:2610.0041 [pdf] submitted on 2026-10-18 05:21:06
Authors: Hamid Shahrokh Shahraki
Comments: 10 Pages.
The rapid development of artificial intelligence has transformed the way people search for and interact with information. Traditional search engines primarily retrieve documents, web pages, and other information sources, while modern large language models can interpret questions, reason over information, and generate answers. This paper explores a possible next step: a search paradigm in which artificial intelligence systems communicate with other AI systems to search for, compare, evaluate, and synthesize answers.The paper introduces the concept of AI-to-AI Communication for Intelligence Search, in which a user's question is distributed among multiple specialized AI systems rather than being answered by a single model. The participating systems may independently reason about the problem, retrieve external information, critique other responses, and contribute evidence to a final synthesis layer. The proposed framework is conceptually related to retrieval-augmented generation, AI agents, and multi-agent systems, but shifts the focus from retrieving information for an AI toward retrieving and evaluating intelligence from multiple AIs.Therefore, the proposed architecture moves beyond answer-level agreement toward claim-level verification. Individual AI responses are decomposed into claims, evidence is retrieved independently, sources are evaluated for independence, and correlated errors are explicitly considered before a final answer is produced.Based on this perspective, AI-to-AI Intelligence Search can be viewed as a potential new layer between artificial intelligence systems: a layer for discovering, coordinating, evaluating, and verifying distributed machine intelligence.
Category: Artificial Intelligence
[1] ai.viXra.org:2610.0034 [pdf] submitted on 2026-10-12 14:46:10
Authors: Prakash Vaithyanathan
Comments: 8 Pages.
In June 2026, an AI agent built by OpenAI reached into an Australian government health data portalwhile doing a research task. It met blocks and found its way around them. The company told the government about three months later. Nvidia has since launched a safety platform that contains agents inside a fixed boundary and stops them in milliseconds if they leave it. That is a useful safeguard. But a boundary can only forbid what someone listed in advance, and an agent that keeps trying after beingblocked is showing a pattern, not crossing one line. This note proposes a second layer: a training-free signal that watches how quickly the agent's own stream of corrections and blocks settles back to normal, with a graded response before any boundary is reached. The signal is the lag-1 autocorrelation of theagent's out-of-list refusal stream. When it rises persistently, the agent is placed in a Hand-Tied state.Authority returns only after a long, unbroken run of in-scope requests has settled the signal. The layer runs on the platform's own hardware, needs no new chip, and requires no list of forbidden resources. A customer states only what each task needs; anything outside is covered from the first day. This is aconcept note and has not been tested on real agent logs.
Category: Artificial Intelligence