Monitoring

How can you detect prompt injection?

Keep records of what each AI tool is given, says and does. Detection will miss some attempts, and the records also give you a trail afterwards.

Last reviewed against NCSC, OWASP GenAI Security Project and Microsoft Security Response Center. Sources

Factory buildings linked through a hub to a phone showing a fingerprint and shield

The answer

Record enough to see

The NCSC advises keeping sufficient records to spot suspicious activity, which may mean storing the complete prompts and replies as well as the tools and APIs the system used [1]. Errors from those tools can be telling, since they may show an attacker refining a method [1].

The NCSC's secure AI guidelines list logging and monitoring under secure operation and maintenance [7].

Signals

Worth reviewing

Megaphone beside a padlock, shield, firewall, cloud and password field

Odd tool use

Calls the user never asked for, especially anything that sends or fetches.

Two server towers, one on fire, linked by a cycle of arrows to a laptop

Outbound links and images

EchoLeak leaked data through URLs [4], and one of OWASP's examples has a model insert a link that leaks the conversation [2].

Scales weighing a pile of paperwork against an alarm clock and coins

Repeated near-identical failures

A run of similar failed calls can mean someone is iterating on an attack, so review them together rather than one at a time.

Automated help

Classifiers

Microsoft uses Prompt Shields to flag injection attempts before they shape the output [3]. Tools like this reduce noise but do not remove the risk.

The quarterly plan shows where monitoring fits, and the filters question covers what detection tools can and cannot do.

Decide who acts

An alert only helps if someone with authority can switch the tool off or cut its access.

Keep reading

Read next

Three colleagues around a laptop with question marks above them

Questions

Short, sourced answers on prompt injection: what it is, whether it can be fixed, how it differs from SQL injection and what UK teams should do first.

Read the guide Hooded figure at a laptop surrounded by a poisoned email, bank card, bug, alarm and warning triangle

Risks

The risks prompt injection creates for UK organisations: data leaks, unwanted actions and over-trusted tools, with NCSC, OWASP and vendor evidence.

Read the guide Infinity loop with a team, padlock, cloud, gears and an alert envelope

Plan

A practical prompt injection plan for UK organisations: map untrusted content, limit tool access, add approvals, keep logs and test, in order.

Read the guide

How CyPro can help

Need a hand planning for this?

Understanding the change is one job. Working out what it means for your systems, your suppliers and your roadmap is several more. CyPro's consultants can take any of them on.

Security lead advising a board

Fractional CISO for the board

A senior security lead who advises the board on what to fix first, what to accept and what to tell customers, and owns the decision record.

More on cypro.co.uk
Cyber security consultants planning hardening work

Hardening and segmentation

Our consultants put the controls in place that limit the damage when something is missed: firewall rules, network isolation and tighter admin access.

More on cypro.co.uk
Penetration tester probing an application for weaknesses

Penetration testing

A focused test of your systems and the controls around them, so you know they hold before an attacker finds out for you.

More on cypro.co.uk
Scanner checking internet-facing systems for weaknesses

Vulnerability scanning

Find every internet-facing system you run, including the forgotten ones, and see which are still on a vulnerable build.

See the scanning service
Security analysts monitoring systems around the clock

24/7 security monitoring

Our SOC analysts watch your systems day and night for the signs of an attack in progress: unexpected processes, new outbound connections, logins that do not fit.

More on cypro.co.uk
Incident responders containing an attack

Incident response

If you may already be compromised, our responders contain it, preserve the evidence, find out what was touched and get you running again safely.

More on cypro.co.uk
Talk to us Book a Free Consultation
Rocket launching from a cloud

Plan ahead

Want a plan for your organisation?

Prompt injection only shows up when someone tries it against your own tools. Penetration testing can probe the AI features you run or build, to see what a hostile instruction could reach.

Test your AI tools Book a Free Consultation