Reading the evidence
Are AI powered attacks a new kind of threat?
In part. Much of the activity is familiar, such as phishing and fraud, but AI changes speed, scale and the targets, because AI tools are now attacked directly.
Last reviewed against NCSC (UK), Anthropic and UK Government (DSIT and Home Office). Sources
Familiar
What has not changed
Phishing still tops the UK list, reaching 38% of businesses in the 2025/2026 survey [6]. The survey does not say how much involves AI, and the Arup fraud was a payment scam of a familiar kind carried out with synthetic voices and images [8]. The routes in, and the controls that block them, are largely the established ones.
New
What AI adds
Effort and speed
Wider access
As AI-enabled cyber tools circulate, more people are likely to be able to carry out intrusions, according to the NCSC [1].
Reading the numbers
A caution
The government survey measures breaches and attacks of every kind, and it does not separate out those that used AI [6]. The figure for businesses, 43%, is therefore a picture of the general threat, not of an AI-specific wave [6]. Anthropic's campaign is a single reported case described by the company itself [5]. Reasonable readers can disagree about how far either points to a lasting change.
What follows
Priorities
Keep the basics strong, shorten the time to fix, add second checks to payments, and put AI tools under the same governance as other systems. The NCSC's own guidance on AI points leaders to security as a core requirement across a system's lifecycle [4]. See the defence plan.
Keep reading
Read next
Can it be fixed?
The NCSC says prompt injection probably cannot be fully prevented. What that means for buying AI products and for how UK teams design them.
Read the guide
Am I exposed?
A practical checklist to see whether your payment processes, inboxes, systems and AI tools are open to AI powered attacks, for UK organisations.
Read the guide
How to defend
An ordered plan for UK organisations to defend against AI powered attacks: payment checks, phishing, patching, AI tool limits, logging and the board.
Read the guideHow CyPro can help
Need a hand with any of this?
Patching is one job. Keeping a vulnerable system safe until it is patched, and proving it, is several more. CyPro's consultants, testers and SOC analysts can take any of them off your plate.
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
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
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
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
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
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
Know your exposure
Not sure you are protected?
Many of the risks here sit in how an organisation uses AI: what tools staff run, what they can reach and who owns the risk. AI governance work sets that out so controls match the tools in use.