Financial services security in 2027: Five areas worth reviewing now

Financial services security in 2027: Five areas worth reviewing now

Financial institutions are no strangers to security assessments, regulatory requirements and resilience planning. Most have established controls, documented procedures and teams responsible for managing different areas of risk.

But having these arrangements in place and knowing how well they would stand up to a serious incident are two very different things.

An organisation might have recently completed a penetration test, reviewed its suppliers and updated its incident response plan. Each exercise provides useful assurance but would those findings tell you what might happen if an attacker compromised a supplier account, gained access to a critical system and disrupted an important business service? More importantly, would the organisation recognise what was happening quickly enough to respond?

As financial institutions plan their security budgets and priorities for 2027, these are the kinds of questions we would recommend asking.

With new FCA operational incident and material third-party reporting requirements due to take effect in March 2027, alongside the continued adoption of AI and changes in how financial services are delivered, there are several reasons to revisit existing security and resilience arrangements.

Here are five areas we think deserve the most attention.

1. Has your security kept pace with changes to the business? 

Technology is changing faster than ever. Organisations are under pressure to adopt AI to keep up with competitors, drive efficiency and increase productivity. New technology, acquisitions, changes in suppliers and the introduction of new services or AI functionality can all affect an organisation’s exposure.

Even relatively small changes can have consequences. A new third-party integration might introduce additional access to internal systems. An employee moving between departments could retain permissions they no longer need. A system introduced for a specific purpose might gradually become critical to day-to-day operations.

None of these necessarily represents a serious weakness on its own. The concern is whether the organisation has recognised the changes and considered their implications.

This is why we believe cyber security maturity assessments need to look at more than the presence of controls or compliance with a particular standard.

They should examine how security works in practice, where weaknesses exist and how those weaknesses could be exploited by an attacker.

For financial institutions planning their security priorities for 2027, a useful starting point is to ask whether their current understanding of risk still reflects how the business operates today.

2. Would your incident response plan work under pressure?

Most financial institutions have an incident response plan, but from our experience, far fewer can be certain how effectively it would work during a serious incident.

Consider a ransomware attack affecting a critical service. The technical team is investigating, senior leadership wants to understand the business impact, customers may be affected and decisions need to be made about recovery, communications and regulatory reporting.

What happens if the person who normally coordinates the response is on leave? If key decision-makers are working remotely, how will they receive accurate information and make decisions quickly? What if the incident involves a third-party provider? Who is responsible for coordinating the response with them?

Teams should also know how they will preserve evidence and coordinate communications if their usual collaboration tools are unavailable or cannot be trusted.

These are difficult questions to resolve for the first time in the middle of a crisis.

The FCA’s new operational incident reporting requirements, due to come into force on 18 March 2027, make this a particularly relevant area for firms within scope. Organisations will need to be able to assess incidents against the applicable reporting thresholds and meet the required notification timescales.

A plan can look perfectly reasonable on paper while leaving important questions unanswered about responsibilities, escalation and decision-making. Tabletop exercises and incident simulations give organisations an opportunity to work through those issues. They can expose assumptions that have never been challenged, particularly around who makes decisions and how different teams coordinate their response. Exercises should test not only who can activate the response and make decisions, but who acts as deputy and what evidence is required before recovery or incident closure is approved.

The value comes from finding those gaps before a real incident forces the organisation to confront them.

3. Do you really know how AI is being used across your organisation?

AI is already being used across financial services, sometimes through formal projects and sometimes through individual employees looking for more efficient ways to work.

The challenge for security teams is that not all of this activity is necessarily visible.

An employee might upload sensitive or regulated information to an external AI tool to analyse a document. A department could introduce an AI-enabled service without fully understanding how information is processed or an existing technology supplier might add AI capabilities to a platform the organisation already uses.

Where AI functionality is being developed or integrated into an existing service, security teams should be involved early enough to understand the architecture, data flows, access permissions and testing completed before it goes live.

How confident are you that your organisation knows where AI is being used and what information those tools can access?

There are legitimate opportunities to improve efficiency and services through AI, but organisations need to understand the security implications.

That includes knowing what data is being shared, how AI-enabled systems are connected to existing infrastructure and whether appropriate governance arrangements are in place.

Organisations should also consider whether employees or customers are allowing AI agents to access accounts, systems or credentials directly, and what controls exist around that access.

It also means considering how AI could be used against the organisation, including more convincing phishing attempts and social engineering attacks.

For 2027, we would encourage financial institutions to look beyond whether they have an AI policy and focus on the more useful question – whether that policy reflects what is actually happening across the business.

4. Are you testing your security the way an attacker would?

Penetration testing, vulnerability assessments and security audits all have an important role in identifying weaknesses.

But attackers do not necessarily approach an organisation in the same way that security assessments are commissioned.

A cyber penetration test might focus on a defined set of systems. A physical security review might examine access controls at a particular site. Employee awareness might be assessed through a separate phishing exercise.

Each assessment could be completed successfully, with findings addressed by the relevant team.

But what happens when an attacker combines those methods?

An attacker might gather information from social media to impersonate a trusted contact, use social engineering to gain access to a building and then exploit an internal system that would be difficult to compromise remotely.

These are the kinds of scenarios we consider through Red Team exercises.

Rather than examining individual vulnerabilities in isolation, the objective is to understand how an attacker could use different methods to achieve a specific goal.

Red Team testing can also examine how effectively an organisation detects and responds to suspicious activity as an attack develops.

For financial institutions, the question is whether their current testing programme provides a realistic understanding of how an attacker might target the organisation, or whether important gaps remain between individual assessments.

5. Who is looking at the risks that fall between teams?

Financial institutions need specialist teams to manage cyber security, operational resilience, compliance, physical security and third party risk.

The challenge is making sure those teams are looking at the same picture.

Take a supplier with access to an important internal system. The supplier assurance team may have completed its checks. The cyber security team may have tested the system. The operational resilience team may have assessed the consequences of a service outage.

But has anyone considered what would happen if the supplier’s account were compromised and used to access that system?

Would the activity be detected? Who would take responsibility for the incident? How quickly could access be restricted without disrupting the service?

Supplier incidents can also create several parallel questions: what data or credentials may have been exposed, whether data processed by the service can be trusted, how operations are restored and whether regulatory notification is required.

These questions cross several areas of responsibility, which can make them easy to overlook.

Through our work across cyber and physical security, we’ve seen how weaknesses that appear relatively insignificant on their own can become much more serious when combined.

That does not mean every organisation needs to restructure its security functions. It does mean there needs to be a way of bringing different findings together and understanding their combined implications.

As organisations review their priorities for 2027, it is worth asking who has responsibility for identifying and addressing the risks that do not fit neatly within one team’s remit.

Looking ahead to 2027

There will always be new threats, technologies and regulatory requirements for financial institutions to consider. But responding to those developments should not come at the expense of understanding how well existing security arrangements work.

As financial institutions plan for 2027, a useful starting point is to consider what has changed across the organisation, how well existing security arrangements have been tested and where assumptions may have gone unchallenged.

The priority should be understanding where those arrangements could fall short, what the consequences might be and what needs to change before an incident exposes the weaknesses.

How Toro supports UK Finance members

As a commercial partner of UK Finance, Toro Solutions provides members with specialist support across threat-led cyber security maturity assessments, operational resilience and incident response, AI risk and Red Team testing.

We help organisations examine how their security arrangements work in practice, identify weaknesses and make informed decisions about where to focus their efforts.

Find out more about the support available to UK Finance members.