Emerging threats

We support organisations striving to build a trustworthy, safe online environment where users can engage authentically in their communities.
Commercial organisationsWe support commercial organisations operating in a digital world, seeking to protect their reputation and prevent business disruption caused by cyber attacks and compliance breaches.
International programmes and developmentWe support international government organisations and NGOs working to provide infrastructure or improve the capabilities, security and resilience of their nation.
UK government and public sectorWe support UK government organisations responsible for safeguarding critical infrastructure, preserving public trust, and maintaining national security.



A new Five Eyes advisory published this week confirms what we're seeing in client environments: the time available to respond to a known vulnerability is shrinking fast. The advisory's message is urgent; AI-assisted attacks are scaling faster, exploits are developing more quickly and the gap between a vulnerability becoming known and being actively weaponised has narrowed dramatically.
The recommended mitigations aren't new, but the time you have to act on them is a lot less.
Cyber Essentials requires critical patches to be applied within 14 days of release. For many organisations, that deadline is already a stretch. But in the current threat landscape, 14 days is no longer a target to aim for—it should be the worst case you should allow yourself.
We're now working in an environment where the realistic window between patch release and active exploitation can be measured in hours, not weeks. The traditional approach of testing patches in a staging environment before rolling them out broadly, which has been sound practice for years, is increasingly untenable. By the time testing is complete, the vulnerability is already being exploited.
Ultimately, Cyber Essentials sets a floor, not a standard. Meeting the 14-day requirement is compliance. Treating it as your target is risk. Your patching posture should be driven by the current threat environment, not the minimum regulatory expectation.
A faster threat environment doesn’t mean abandoning all testing, but it does mean rethinking the model. Here are some practical steps:
Prioritise. And do so, ruthlessly. Not all patches carry equal risk, so focus your fastest deployment on internet-facing systems, edge devices and anything in your critical infrastructure. Firewalls and VPN appliances in particular are a primary entry point for current threat actors.
Compress your testing window, not your coverage. Where a staging test is necessary, limit it to hours, not days. Accept that some residual risk in the testing process is lower than the risk of a delayed rollout.
Plan for the gaps. Faster patching is essential, but it’s not a complete strategy. Reduce your attack surface, limit lateral movement within your environment and ensure that what is externally exposed is as small as possible. A vulnerability you cannot patch immediately should not give an attacker a path to everything.
Review your legacy tech. Systems that can’t be patched promptly (or at all) carry disproportionate risk in the current environment. Identify them, isolate them where possible and plan for their replacement.
The Five Eyes advisory is clear that AI is accelerating the pace of attacks at scale. Phishing campaigns are now customised per target and deployed in volume. Known vulnerabilities—particularly on Cisco and Fortinet edge devices—are being exploited faster than ever. The advice to patch, reduce legacy exposure and prepare proactively has not changed. The urgency behind it has.
If you would like to discuss your current patching posture, review your exposure or understand how your existing controls measure up against the current threat landscape, let’s talk.

Sector-specific security training helps organisations move beyond abstract awareness and develop a concrete, practical understanding of the risks they actually face.

Protection Group International has become a founding signatory of the CREST AI Charter – a global initiative establishing professional standards for responsible AI use across the cybersecurity industry.

HMRC has just signed a £175 million, ten-year contract with Quantexa, a UK-based data, analytics and AI software company.