This post covers what two official bodies say defenders should watch toward 2027. The sources are the UK National Cyber Security Centre (NCSC) assessment of the impact of AI on cyber threat from now to 2027, published 7 May 2025, and the ENISA Threat Landscape 2026, dated September 2026. Each priority is a question drawn from a stated finding, not our own forecast.
What do the NCSC and ENISA say about the years ahead?
The NCSC says its paper uses the Professional Head of Intelligence Assessment probability yardstick every time it makes an assessment, judgment or prediction. So terms such as “almost certainly” and “highly likely” below are the NCSC’s own, kept as written. The paper focuses on the use of AI in cyber intrusion, and the NCSC says that AI and its application to cyber operations is changing fast and that “Technical surprise is likely.”
ENISA’s report analyzed 8 257 incidents recorded from 1 January 2025 to 31 December 2025. ENISA assesses that several of the dynamics it observed during 2025 are likely to persist into the next reporting period. Its outlook is based on that 2025 data. The two papers differ in date, so read each against its own.
Will the time between disclosure and exploitation keep shrinking?
Exploitation timing is a main theme of the NCSC paper. The NCSC assesses that the time between disclosure and exploitation has shrunk to days and that AI will almost certainly reduce this further. It also assesses that by 2027 AI-enabled tools will almost certainly enhance threat actors’ capability to exploit known vulnerabilities, increasing the volume of attacks against systems that have not been updated with security fixes.
The NCSC adds that the most significant AI cyber development will highly likely come from AI-assisted vulnerability research and exploit development. For skilled actors able to fine-tune AI models, it assesses that AI will highly likely enhance zero-day discovery and exploitation to 2027. It defines a zero-day as a vulnerability with no security fix available.
ENISA gives a baseline for today. It reports that exploitation of vulnerabilities, including N-day and 0-day, remained a prevalent intrusion vector (60.4%), and that CISA’s Known Exploited Vulnerabilities catalog added 245 vulnerabilities over the reporting period. See our post on exploited vulnerabilities in the ENISA report and the glossary entry on known exploited vulnerabilities.
| Topic | NCSC | ENISA |
|---|---|---|
| Document | Impact of AI on cyber threat from now to 2027, published 7 May 2025 | ENISA Threat Landscape 2026, September 2026 |
| Period covered | Assessment of impacts between now and 2027 | Incidents from 1 January 2025 to 31 December 2025 |
| AI | Fully automated end-to-end advanced attacks unlikely to 2027 | Likely 2026 sees more kill chain phases directly enabled by AI |
| Vulnerabilities | Disclosure to exploitation has shrunk to days | Exploitation of vulnerabilities seen in 60.4% of cases |
| Dependencies | AI in CNI almost certainly presents an increased attack surface | Software suppliers, cloud platforms and service providers remain high-value targets |
How far does AI automation go by 2027?
The NCSC is specific about limits. It assesses that, to 2027, AI will highly likely increase the volume and impact of intrusions through evolution and enhancement of existing tactics, rather than creating novel threat vectors. It assesses that fully automated, end-to-end advanced cyber attacks are unlikely to 2027, and that skilled actors will need to remain in the loop. Skilled actors will almost certainly keep experimenting with automating parts of the attack chain.
ENISA sets a nearer horizon. It assesses that AI will highly likely continue to act mainly as a force multiplier, and that it is also likely 2026 will see an increased number of the kill chain’s phases directly enabled by AI, with possible experimentation of human-out-of-the-loop proofs of concept. ENISA says this would be likely to hinder detection and imputation for network defenders. Our post on what agencies report on AI cyber threats covers the 2026 reporting in more detail.
What does the attack surface of AI systems mean for defenders?
The NCSC assesses that the growing incorporation of AI models and systems across the UK’s technology base, particularly within critical national infrastructure, almost certainly presents an increased attack surface. It says techniques such as direct prompt injection, software vulnerabilities, indirect prompt injection and supply chain attack are already capable of enabling exploitation of AI systems. It also assesses that organizations using AI systems will almost certainly need to keep security measures on those systems and their dependencies up to date.
ENISA says it is also likely that use of new AI models will increase EU organizations’ exposure to new threats, arising from both accidental and malicious activities.
Which findings that are not about AI belong on the list?
ENISA’s outlook is broad. It says the practical differences between cybercrime, hacktivist and State-nexus activity are likely to further converge, with the same vulnerabilities, platforms and toolsets leveraged regardless of motivation. It says software suppliers, developers, managed service providers, cloud platforms and customer-service environments continued to appear as high-value targets, and that device-code phishing and ClickFix illustrate a shift towards abusing trust rather than technical weaknesses alone. On operational technology, ENISA says hacktivist claims will likely increase while confirmed incidents with operational impact are likely to remain stable.
The NCSC Annual Review 2025 chapter adds a measured fact: its incident management team faced a record number of nationally significant incidents. See our posts on the NCSC Annual Review threat findings and lessons from official incident reports. For a separate planning thread, see post-quantum migration milestones from the NCSC and the EU.
What questions can a team ask from these findings?
The checklist below turns stated findings into questions. It is not advice from the agencies, and it does not promise any outcome.
- Patch timingThe NCSC says disclosure to exploitation has shrunk to days. How long do your fixes take?
- Known exploited flawsENISA cites 245 additions to the CISA catalog. Do you check it against your systems?
- AI systemsThe NCSC names prompt injection and supply chain attack. Where is AI connected to your data?
- DependenciesENISA names suppliers and cloud platforms as high-value targets. Which ones would hurt you most?
- Trust abuseENISA names device-code phishing and ClickFix. Do staff know these tricks?
Last checked against the official pages on 9 October 2026. Dates and status can change, so confirm on the pages listed under Sources before acting.
For background on how threats are cataloged, see the Brandefense knowledge hub pages on threat groups, malware families, ATT&CK techniques and significant incidents, and the glossary. The cyber threat intelligence page describes the related service.