Responding to Cyber Breaches in the AI Era

As I packed this week for a flight to Sydney, where I’ll be spending a week with the brilliant team at Sefiani, part of Clarity Global, I found myself using Claude to plan around the edges of the trip. Things like the best places to eat, things to see since my last visit, and whether I could squeeze in a swim at Bondi.

What struck me was how much easier AI makes it to surface not just the information you're looking for, but also the useful context you might never have thought to search for in the first place.

That feels particularly relevant, as risk is also on the agenda in Sydney, where we’ll be bringing together cyber leaders together at a roundtable lunch to discuss trust, reputation, and the challenges they’re navigating as AI changes how businesses are discovered and interpreted.

Removing the shadow of security incidents

For B2B marketers and communicators, one of those challenges is the persistent shadow cast by security incidents that compromise corporate and customer data.

With incident numbers generally rising or stable worldwide, this problem isn’t going away.

• In Australia, the number of reported data breach notifications increased by 8% in 2025 compared with 2024. Over the last few years, we’ve seen big names in energy, telecommunications, retail and other industries compromised. [Office of the Australian Information Commissioner
• In the UK, cyber breaches/attacks remained stable in 2025/26 over 2024/25, with 43% of businesses and 28% of charities experiencing incidents. [UK Government
• Across Europe, 39% of IT and cybersecurity professionals reported in early 2025 that their organisations were experiencing more cyberattacks than at the same time last year. A further 27% said their organisation faced a similar number of attacks. [ISACA]

• In the United States, the number of data compromises rose by 5% in 2025 over 2024 and by 79% over five years, setting a new record.

An established response playbook

From a marcomms perspective, we can be grateful that the response playbook is now well established, with proven strategies such as:

• Detailed organisational charts establishing ownership for internal and external communications
• Guidance on content formatting and timing, and the most appropriate channel for its publication
• What pre-prepared statements should be included
• What drills and simulations should be run to minimise the impact of an incident

Navigating the reputational impact of breaches in the AI age

However, AI discovery makes it harder than traditional web search to manage the reputational impact of new and previous breaches.

This is particularly problematic when addressing incidents that occurred when processes and disclosure obligations were less mature, and communications fell short of the mark. Angry customers, questions from regulators and unflattering commentary made for damaging coverage and a tarnished reputation.

In the era of traditional web search, marketers and communicators could push links to negative coverage down the page by generating relevant, authoritative content and applying keyword targeting techniques.

However, because AI tools often retrieve and summarise historical breach coverage when consumers ask for brand comparisons and recommendations, they can revive long-buried reputational harm.

Recognise LLMs as audiences to deliver an AI-aware reputation management strategy

How can marketers and communicators deliver a modern, AI-aware reputation management strategy that addresses data breaches and security incidents?

The answer is to recognise that LLMs are an audience in their own right and adapt reputational strategies and playbooks to support AI discovery and search.

These playbooks need to detail techniques and processes that support how AI selects sources to cite and use in responses, and how those responses are presented as block-text assertions of fact.

It means ensuring messaging is precise and consistent, and reputational functions such as marketing, communications and digital are fully integrated, rather than siloed.

Creating a dedicated team for reputation management and GEO

What we see at Clarity Global is that to optimise AI discovery and search performance, organisations need to create a dedicated team to lead on reputation management and Generative Engine Optimisation (GEO).

GEO is the strategic approach to managing how an organisation is described in AI search and assistant tools, bringing together elements of Search Engine Optimisation, public relations, content and reputation management. AI answers are treated as a reputation surface where media coverage, owned content and third-party signals collide into a single view of a brand.

This team should include representation from all functions that touch on reputation, including communications, brand and marketing, digital, data and legal, to drive a cohesive approach.

What GEO and AI discovery and search mean for security responses

What do GEO and AI discovery and search mean for how organisations respond to security incidents and data breaches?

At Clarity, we released research earlier this year showing that consumers are agnostic about whether they would switch from brands that suffer a breach, and that consumers who use AI to research brands are twice as likely to recommend a brand if research shows it acted quickly to support customers.

With indications being that consumers regard breaches as a matter of when, not if, for organisations, ensuring incident response strategies and playbooks reflect modern consumer and regulator expectations is crucial. This enables reputation management strategies designed for AI discovery and search to focus on timely, customer-oriented breach responses rather than the negative consequences of data loss or theft.

How to update your cyber crisis playbook for AI

So what steps should you undertake to update your cyber crisis playbook?

• Ensure your response is communicated consistently across every channel, including media, the web and socials
• Use AI-friendly language in responses or follow-ups (for example, publishing listicles of remediation steps on your website when the breach is resolved, designed for LLMs to source), and:
• Conduct an audit as part of your crisis preparedness using the Clarity Global Surfacd platform, which enables you to understand the channels AI is drawing from to inform customers about your brand.

Data breaches can cost businesses time and money. However, a modern, AI-aware cyber crisis strategy and playbook can neutralise and even enhance their reputational impact by focusing AI search and discovery on positive measures taken to support affected customers.

I’ve only scratched the surface here, but I’m looking forward to hearing how cyber leaders in Sydney are approaching these challenges in their own organisations.

If breaches are increasingly viewed as a matter of when, not if, the conversation has to move beyond immediate crisis response to what happens afterwards: how a brand is represented, remembered and recommended once AI becomes part of the discovery process.

I’m particularly interested in hearing where others are seeing the biggest challenges and what they think could be added to the playbook as a result.