| ai.viXra.org > Artificial Intelligence > 2606 |
|
 |
| ai.viXra.org > Artificial Intelligence > 2606 |
|
|
[3] ai.viXra.org:2606.0023 [pdf] submitted on 2026-06-09 20:23:35
Authors: Sudipto Chandra
Comments: 7 Pages. Creative Commons Attribution-NonCommercial 4.0 International
Between 2022 and 2026, commodity bot mitigation shifted from binary, single-signal blocklisting toward composite trust scoring: a request is admitted only if it satisfies a conjunction of independently evaluated detection layers. Cloudflare, which fronts a large share of all sites that deploy bot-mitigation tooling, exemplifies this architecture with a stack we decompose into fifteen layers grouped into five domains — network and transport, client environment, behavioral, interactive challenge, and programmable-cryptographic controls. This paper offers a defender-oriented taxonomy of that stack. For each layer we characterize the signal it reads, the structural reason it is or is not evadable, and the residual hardness that remains after best-effort evasion. Our central observation is that evadability is not uniform: layers that read artifacts a client emits (TLS handshakes, HTTP/2 frame ordering, header order) are structurally weak because the artifact can be reproduced exactly, whereas layers that read properties a client must continuously possess (genuine per-zone behavioral history, a privately held cryptographic credential) resist forgery by construction. We argue that the long-run defensive trajectory follows from this asymmetry, and that newer mechanisms such as cryptographic agent attestation are a different kind of control altogether: forgery-resistant by construction. Their present weakness is not cryptographic but a matter of deployment, since current configurations fail open when no signature is presented.
Category: Artificial Intelligence
[2] ai.viXra.org:2606.0002 [pdf] submitted on 2026-06-01 20:10:29
Authors: Keiji Yoshimura
Comments: 10 Pages.
Generative AI enables non-experts to rapidly produce papers, code, figures, repositories, quality-assurance documents, and submission metadata. This capacity can widen access to knowledge production, but it also creates a risk that artifact completeness is mistaken for domain validity. This manuscript presents a reflexive case study of two AI-assisted research attempts conducted by the author: a lithium-ion battery-model audit and an A10-derived Hubble-tension scaffold audit. In the battery case, an initially technology-oriented project was reduced to a residual-drift audit of simplified PyBaMM models against external discharge time series. Claims about safe charging, practical battery-management-system readiness, degradation reduction, lithium-plating avoidance, and real-cell validation were explicitly prohibited. In the Hubble case, a cosmological scaffold failed to establish a first-principles physical E(z)/H(z) bridge and therefore stopped before real-data likelihoods, MCMC, posterior comparison, or evidence evaluation. Across two distinct domains, the same pattern appeared: AI inflated research artifacts, but strict grounding checks reduced them to claim-boundary and failure-boundary records. This manuscript does not argue that AI-assisted research is useless, nor does it evaluate expert-led AI science. It argues that non-expert AI-assisted research requires explicit domain-grounding, prior-art, external-data, implementation-or-experiment, expert-review-worthiness, and public-claim-boundary gates before public claims are made.
Category: Artificial Intelligence
[1] ai.viXra.org:2606.0001 [pdf] submitted on 2026-06-01 19:06:25
Authors: Stephane H Maes, Andrey Ryabov, Ramu Sunkara
Comments: 9 Pages. All related details of the projects (and updates) can be found and followed at https://shmaes.wordpress.com/
This position paper argues that static tool-registry protocols, most prominently the Model Context Protocol (MCP), are unsuitable as foundational integration architectures for production-grade agentic AI in heterogeneous enterprise environments, and that the field should converge on "Real-Time Discovery and (Self) Coding" (RTDC) Integration as the necessary architectural alternative. MCP was introduced to standardize connections between language models and external systems through a uniform JSON-RPC 2.0 client-server interface. Despite rapid adoption in developer tooling, we demonstrate that MCP doubles the enterprise integration maintenance and vulnerability surfaces, fails categorically against legacy systems that harbor the most critical enterprise data, exhausts language model context windows through static tool enumeration at production scale, induces selection collapse in large tool registries, and introduces security vulnerabilities, including tool poisoning, prompt injection, and supply chain compromise. These are fundamentally incompatible with enterprise zero-trust architectures. We survey three alternative paradigms: user-interface automation, real-time context lake architectures, and terminal agents. We show that each resolves only a subset of the enterprise integration challenge. We then argue for RTDC Integration: a paradigm in which autonomous meta-agents dynamically discover system interfaces, and schemas, at runtime, synthesize and validate integration code in sandboxed execution environments, and accumulate versioned, reusable capability artifacts through a continuous discovery-synthesize-verify-promote loop. Unlike static protocol registries, RTDC integrates with legacy mainframes, undocumented APIs, and proprietary systems without prior engineering effort, and enables the eventual decommissioning of those systems through autonomous logic internalization. This is a capability no protocol-based approach can provide.
Category: Artificial Intelligence