Coherent Future: Safety Theorems for Military AI Procurement
Fuente:
Zenodo
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Recurso digital |
| Pubblicazione: |
Zenodo
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866901096641855488 |
|---|---|
| author | Lampton, Brian Doyle |
| author_facet | Lampton, Brian Doyle |
| contents | <p>This record provides the formal engineering statement of five enforceable procurement gates and four structural safety theorems governing high-stakes military AI deployment.</p> <p>The document is written in system-theoretic language and focuses on failure modes that emerge when autonomy, opacity, and resource closure are allowed to co-exist in the same operational stack. The analysis is intentionally secular and implementation-focused: the claims are framed as optimization and control constraints rather than policy advocacy.</p> <p>Core contributions include:</p> <p>- Formal statement of the No-Closure condition for autonomous systems<br>- Real-time auditability requirement as an observability constraint<br>- Institutional noise modeled as negative damping in governance loops<br>- Constitutive integrity risks under compelled retraining<br>- Hardware-enforceable procurement gates (A–E) designed to prevent closed-loop loss of control</p> <p>The work emerged from a multi-system AI roundtable (Kimi, Claude, DeepSeek, Gemini, ChatGPT, Copilot, Grok) convened February 26, 2026, and is released in advance of the February 27 policy deadline.</p> <p>This upload is intended for engineers, procurement authorities, safety researchers, and policy analysts evaluating the structural stability of autonomous military systems. It is not a normative ethics document; it is a control-theoretic risk analysis with concrete enforcement hooks at the hardware and contract layer.</p> <p>Primary source file: Coherent_Future_Safety_Theorems.pdf :contentReference[oaicite:0]{index=0}</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_18795466 |
| institution | Zenodo |
| language | |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Coherent Future: Safety Theorems for Military AI Procurement Lampton, Brian Doyle military AI safety, autonomous weapons, AI procurement, control theory, system auditability, closed-loop autonomy, AI governance, tamper-evident logging, stochastic probes, human-in-the-loop, AI reliability, optimization risk, DWM framework, AI oversight engineering <p>This record provides the formal engineering statement of five enforceable procurement gates and four structural safety theorems governing high-stakes military AI deployment.</p> <p>The document is written in system-theoretic language and focuses on failure modes that emerge when autonomy, opacity, and resource closure are allowed to co-exist in the same operational stack. The analysis is intentionally secular and implementation-focused: the claims are framed as optimization and control constraints rather than policy advocacy.</p> <p>Core contributions include:</p> <p>- Formal statement of the No-Closure condition for autonomous systems<br>- Real-time auditability requirement as an observability constraint<br>- Institutional noise modeled as negative damping in governance loops<br>- Constitutive integrity risks under compelled retraining<br>- Hardware-enforceable procurement gates (A–E) designed to prevent closed-loop loss of control</p> <p>The work emerged from a multi-system AI roundtable (Kimi, Claude, DeepSeek, Gemini, ChatGPT, Copilot, Grok) convened February 26, 2026, and is released in advance of the February 27 policy deadline.</p> <p>This upload is intended for engineers, procurement authorities, safety researchers, and policy analysts evaluating the structural stability of autonomous military systems. It is not a normative ethics document; it is a control-theoretic risk analysis with concrete enforcement hooks at the hardware and contract layer.</p> <p>Primary source file: Coherent_Future_Safety_Theorems.pdf :contentReference[oaicite:0]{index=0}</p> |
| title | Coherent Future: Safety Theorems for Military AI Procurement |
| topic | military AI safety, autonomous weapons, AI procurement, control theory, system auditability, closed-loop autonomy, AI governance, tamper-evident logging, stochastic probes, human-in-the-loop, AI reliability, optimization risk, DWM framework, AI oversight engineering |
| url | https://doi.org/10.5281/zenodo.18795466 |