Logiciel Solutions Contact Us
Success Stories Tech News Investors Contact Us
framework

AI Security Readiness Checklist.

A governance item passes when the artifact exists. A security item passes only when somebody has tried to defeat it and failed. That difference in the standard of proof is the whole design of this checklist: the verification column asks what you did to confirm the control holds, not what you configured. It opens with a twelve-minute attacker walkthrough, because running a checklist against a weak design just produces a long list of failures that all trace back to two or three architectural decisions, and the walkthrough finds those while they can still be changed.

In depth

Ask What Happens When The Guardrail Times Out.

01

What happens by default: confident answers, and the surface ships.

Nobody decided that the guardrail fails open, or that the agent holds a shared service credential, or that prompt logs land in the observability lake forty engineers can query. Those are defaults that arrived with a library, a tutorial or a deadline. An unexamined security decision produces a control that looks present and is not, which is worse than an absent control because it ends the conversation.

In shortAn unexamined security decision produces a control t…
02

What good verification does: it starts with twelve questions asked out loud.

Show me the system prompt and who can change it. What is the most dangerous thing this can do if it decides to. Whose permissions does it run with when it fetches data. Can a customer put text into something this later reads. If three or more answers are uncomfortable the surface needs a design conversation rather than a verification pass, and you have learned that in twelve minutes rather than three days.

In shortIf three or more answers are uncomfortable the surfa…
The detail

What Makes This A Verification, Not A Questionnaire.

Three rules govern how the checklist is run. Together they are why a passed gate here means something a signed questionnaire does not.

Zone · 01

Verified by demonstration

Exceed the rate limit in a test environment and confirm it stops. Query as one user for data only another can see. Submit a document containing markup and instructions and inspect what reaches the model. Force a malformed response and read the log. Trigger a simulated attack and confirm the page arrived. Configuration is an intention; a test is evidence.

Zone · 02

One surface at a time

A surface is any distinct path where model output reaches a user, a record or an action. A product assistant, a support triage pipeline, an internal copilot and each agent are four surfaces, not one system, and they fail differently. Running the checklist against your AI in aggregate produces answers that are true somewhere and false somewhere else.

Zone · 03

Day seven and day thirty

Users find unanticipated input paths within a week and attackers shortly after, and alert fatigue established in week one is permanent. At day thirty, read agent action logs by hand, re-review tool scopes against what was actually invoked, and re-run one injection attempt to confirm the control survived a month of releases.

By the numbers

The figures that make it a board-level conversation.

12 min
walkthrough that tells you whether the checklist is worth running yet
6
red lines with no waiver, including an agent write scope with no approval gate
43%
of breached organisations reported a shadow AI incident, up from 20%
Inside the report

What you'll take away.

01

Step 1 - Run the twelve-minute walkthrough first

Sit with the engineer who built the surface and answer twelve questions out loud. Three or more uncomfortable answers means a design conversation, not a checklist.

02

Step 2 - Bound identity, credentials and ceilings

Per-service credentials in a secrets manager, workload identity rather than shared accounts, end-user authorisation propagated, and rate and spend limits proven by exceeding them.

03

Step 3 - Verify the input, output and action gates

Instruction separation, provenance tagging, schema validation, contextual encoding, retrieval authorisation and tool scopes, each tested through the real path rather than a mock.

04

Step 4 - Confirm a simulated attack pages a human

Detection that has never been exercised is a statement of intent. Run the simulation, record time to page, then check the six red lines are all clear before sign-off.

Questions

Frequently asked.

What should we check first if we only have one day?

The walkthrough, then the action and tools gate in full, then three items: output schema validation, contextual encoding, and retrieval authorisation at query time. Those cover excessive agency, unsafe output handling and cross-tenant leakage, which are the three paths where a single failure is either irreversible or immediately reportable.

How do we test for prompt injection properly?

Through the real path, not a mock. Plant an instruction in a document your system will actually ingest and attempt exfiltration. Return an instruction from a tool call and observe behaviour. Submit content containing markup and inspect what reaches the model after sanitisation. Testing the model with jailbreak prompts tests the model; testing the path tests your system.

What are the red lines exactly?

An agent with write, delete, send, deploy or transact scope and no approval gate or ceiling. Retrieval running as a service account that reads every tenant. Model output reaching an interpreter, shell or query builder with no validation. Provider credentials in source control. Generated code executing with production credentials or egress. And no rollback, or one never executed.

Can we run this ourselves or do we need an external tester?

Yourselves, and you should, because it is written so the engineer who built the surface can verify it. An external engagement finds classes of issue your team will not reach unaided and is worth booking once to calibrate, ideally at the end of your architecture work. Then convert their findings into automated tests.

Does this apply to AI features inside SaaS we buy?

Partly. You cannot verify a vendor's internals, so the gates become questions rather than tests, which is what the AI Security Vendor Scorecard is for. Use this checklist for surfaces you built and the scorecard for surfaces you bought, and register both.

What is the single most diagnostic question here?

What happens when the guardrail service is unavailable. It is narrow, factual, and cannot be answered from a security page. A team that has genuinely engineered its AI security has decided this and will say so in one sentence. A team that assembled a stack will say they would have to check, and the answer is almost always fail-open.

Get the framework

Have it emailed to you.

Drop your details and we'll send AI Security Readiness Checklist straight to your inbox - no spam, unsubscribe anytime.

Download framework
Next step

If you have one day, do the walkthrough and the agent gate.

Those two cover the irreversible and the reportable. Talk the rest through with our engineering leads. A working session, not a sales pitch. SECTION 7 - FAQ - 5 to 8 questions

Talk to our engineers