In questo argomento
Generative AI · Agents · RAG · Evaluation · Model security · Governance
Deploy useful AI with appropriate controls
How should you evaluate an AI assistant or agent before deployment? Define a task, an authorised corpus and observable failure criteria. Then measure accuracy, source quality, access control, cost and response time on representative cases. The decision record preserves model, instruction and document versions. It distinguishes response quality from permission to execute an action.
Leggi la guida ENMonitoraggio delle notizie
Tutte le notizie internazionali →Titoli salvati · aggiornamento non disponibile · Ultimo recupero riuscito : 22 ott 2026, 21:24
AI Security nei cloud sovrani e privati: TrendAI e NVIDIA uniscono le forze per proteggere l'AI - ImpresaCity
via Google News · aggregatorAI, conto salato per le big tech: nel 2027 bruceranno cassa per 163 miliardi - Il Sole 24 ORE
via Google News · aggregatorPutting Wonder on Trial: Magic vs. Artificial Intelligence - Il Messaggero
via Google News · aggregatorAI security: TrendAI e NVIDIA per il cloud sovrano - bitmat.it
via Google News · aggregatorAI security: TrendAI e NVIDIA per il cloud sovrano - LineaEDP
via Google News · aggregatorRiferimenti: dati, usi e limiti
Documentazione consultata il .
Le note dettagliate di questa edizione sono disponibili in inglese.
| Fonte e finalità | Dato o riferimento verificato | Come utilizzarlo | Limite da conservare |
|---|---|---|---|
| MITRE ATLAS Examine documented tactics, techniques and cases involving AI systems. |
The official repository publishes ATLAS tactics, techniques and case-study data. They support analysis of the actual deployed system. | Select scenarios matching your application’s data, interfaces and tools. Test permissions and consequences, not only model responses. | A technique catalogue is neither an exhaustive coverage guarantee nor a test result for your system. |
| NIST · AI Risk Management Framework Organise management of AI-related risks. |
The AI RMF Core is organised around Govern, Map, Measure and Manage. It is a voluntary risk-management framework. | Define intended use, affected people and error consequences. Connect each risk to an evaluation, owner and stop or recovery mechanism. | The framework certifies neither a model nor an application. Using it does not remove the need to examine applicable requirements. |
| OWASP · LLM Top 10 Prepare tests for language-model application risks. |
The 2025 edition includes Prompt Injection, Sensitive Information Disclosure and Excessive Agency. | Turn a risk into an observable scenario: hostile document, restricted data or out-of-scope tool. Record expected behaviour and evidence of refusal. | A risk list is not a complete assessment. Input filtering alone does not establish authorisation for every action. |
RAG: diritti, citazioni e cache
Le note dettagliate di questa edizione sono disponibili in inglese.
RAG does not delegate permissions to the model
The corpus supplies information; the access service decides which passages an identity may receive. Distinguish relevance selection, object permission and authority to act after reading.
- Cover vector retrieval, lexical branches, citations and caches.
- Apply rights removal to existing conversations.
- Verify answer quality and absence of disclosure separately.
Explore the RAG access matrix and replay cases →
References consulted on · OWASP · LLM08:2025 Vector and Embedding Weaknesses OWASP · API1:2023 Broken Object Level Authorization
OAuth/OIDC: destinatari e accesso
Le note dettagliate di questa edizione sono disponibili in inglese.
An AI agent retains each tool’s authorisation boundaries
A model instruction does not create a grant. For every tool, identify the calling client, resource server, audience and object policy.
- Restrict permissions to the operation and recipient needed for the use case.
- Do not forward one tool’s token to another API merely for convenience.
- Test that an agent cannot read another tenant’s objects despite having a read scope.
See token recipients and the ten OAuth/OIDC cases →
Sources reviewed on · IETF · RFC 9700 · OAuth 2.0 Security BCP · IETF · RFC 9068 · JWT access token profile · OpenID Foundation · OpenID Connect Core 1.0 · IETF · RFC 8725 · JWT Best Current Practices · IETF · RFC 6749 · OAuth 2.0 framework
SSRF: URL e connessioni
Le note dettagliate di questa edizione sono disponibili in inglese.
An AI agent tool keeps bounded network access
Text produced by a model, retrieved document or search result must not expand an allowed tool destination. Defining the connector by operation and service makes the agent’s actual network reach controllable.
- Reconstruct requests from business identifiers when the service is known.
- Enforce resolution, connection, proxy and each redirect at the egress component.
- Keep a traceable decision and response limits, including after retry or fallback.
Read the 12 cases and connection controls →
Sources reviewed on · OWASP · SSRF Prevention Cheat Sheet · WHATWG · URL Living Standard · AWS EC2 · Instance metadata options · IETF · RFC 3849
IA: verificare un risultato JSON
Le note dettagliate di questa edizione sono disponibili in inglese.
An extraction contract precedes every agent action
An AI-generated object remains an input to validate. Correct types establish neither document values, actor identity nor permission to write into the target system.
- Validate fields before handing them to the connector.
- Route unknown values and missing evidence into an explicit review path.
- Apply context and authorization controls to every downstream operation.
Compare the 14 outputs and their reasons →
Sources reviewed on · IETF · RFC 8259 · JSON Schema · Object reference · JSON Schema · Draft 2020-12 validation · Python 3.12 · json decoder · python-jsonschema · Schema validation · OWASP · LLM05:2025 Improper Output Handling