Timeline
Guidance from the NCSC, government statistics, a company's own report and two surveys make up the record. Few of them measure AI-assisted attacks directly.
Last reviewed against NCSC (UK), UK Government (DSIT and Home Office) and Anthropic. Sources
Reading the record
NCSC pages on AI and on phishing both carry the date 13 February 2024 [9]. Over the following two years the record adds government policy on securing AI [6], the NCSC's outlook to 2027 [1], an AI company's account of criminals using its tools [5] and two industry surveys [7] [8]. The most recent entry is the government's Cyber Security Breaches Survey of 30 April 2026 [4].
Entries use the date shown on the source page itself. Where a page gives less than a full date, the entry shows only what the page gives.
Go back to the overview or on to the defence plan.
Timeline
Newest first. Each entry links to its source.
Latest
It found that 43% of businesses had a breach or attack in the past 12 months and that phishing reached 38% of businesses. It also recorded how many organisations had practices to manage AI-related risks.
An NCSC blog by Dave Chismon argued that prompt injection is not SQL injection and that the aim should be to reduce its likelihood and impact.
The Register reported that 62% of security leaders surveyed had seen AI-based attacks on staff in the past year, including fake audio and video and prompt injection.
Anthropic described a data extortion campaign against at least 17 organisations and North Korean operatives using AI to obtain remote technical jobs. It said AI has lowered the barriers to sophisticated cybercrime.
The NCSC judged that AI will help attackers work faster and more efficiently on some intrusion steps. It also noted that attackers already use AI to help gain entry through social engineering.
The code sets baseline cyber security principles for securing AI systems and the organisations that develop and deploy them.
Tech Monitor reported that 54% of UK businesses had no response plan for AI-driven phishing and that only 33% of employees identified fraudulent communications.
The NCSC's pages on AI and cyber security and on defending against phishing carry this date. The first describes prompt injection as one of the most widely reported weaknesses in large language models.
Key dates
7 May 2025
NCSC publishes its assessment of AI and the cyber threat to 2027[1]
27 August 2025
Anthropic publishes its August 2025 threat intelligence report[5]
23 September 2025
Gartner survey on AI-based attacks against staff reported[7]
8 December 2025
NCSC blog sets out why prompt injection is not like SQL injection[2]
30 April 2026
Cyber Security Breaches Survey 2025/2026 published[4]
2027
Year the NCSC's assessment of AI and the cyber threat looks ahead to[1]
Keep reading
Sourced answers on AI phishing, voice scams, prompt injection and what UK organisations should do, from the NCSC, government and published research.
Read the guide
A checklist for UK organisations: which requests, teams and AI tools could be manipulated by AI-assisted phishing, impersonation or prompt injection.
Read the guide
An ordered plan for UK organisations: layered phishing defences, second-channel checks, blame-free reporting and limits on your own AI tools.
Read the guideHow CyPro can help
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.
Find every internet-facing system you run, including the forgotten ones, and see which are still on a vulnerable build.
See the scanning service
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
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
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
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
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
Defending against AI-assisted social engineering is a matter of decisions: who may approve what, how requests are confirmed and what AI tools may do. A virtual CISO helps leaders set those rules and have them tested.
