AI is transforming how organisations create, access and use data. As adoption grows, so does the need to keep sensitive information protected and under control.
In CyberOne’s recent webinar, Data Security in an AI-Driven World, Luke Elston, Microsoft Practice Director at CyberOne, explored why data security and AI security must now be treated as one connected challenge. The session focused on the practical steps organisations can take to reduce risk, strengthen governance and use AI securely.
Many organisations are already struggling to answer basic data security questions: what sensitive data they hold, where it is stored, who has access to it and how it is protected. AI does not remove these challenges. It makes them more urgent.
Luke highlighted that AI increases the speed and scale at which data is accessed, summarised and shared. Weak permissions or governance controls are exposed more quickly, making it essential to get the basics right.
One of the biggest risks discussed was the use of unmanaged AI tools by employees. A user may upload a customer contract, financial record or internal document into a public AI tool simply to summarise it or work more efficiently. There may be no malicious intent, but sensitive data has still left the organisation
This is why user education, AI governance and technical controls are essential. Organisations need clear policies on approved AI tools, combined with controls that prevent sensitive data from being copied or uploaded into unmanaged platforms.
The webinar highlighted the importance of classifying data so organisations can clearly identify whether information is public, internal, confidential or highly confidential. Without classification, it becomes much harder to apply the right protection. Once data is labelled correctly, organisations can connect those labels to controls such as Data Loss Prevention, encryption and retention policies. This allows security teams to move from guesswork to clear, enforceable policy.
A key part of the session focused on Microsoft Purview and how it strengthens data security. Luke outlined how Purview can help organisations:
The session also covered core Purview capabilities: Data Loss Prevention, Information Protection, Insider Risk Management, Compliance Manager and Data Lifecycle Management.
The webinar also explored the relationship between Microsoft Copilot and Purview. Copilot follows existing permissions in the Microsoft tenant, so users only see information they are authorised to access. If permissions are too broad, Copilot can surface sensitive data to users who technically have access but should not.
This reinforces a key message from the session: securing AI starts with securing the data, identities and permissions behind it.
While the webinar focused heavily on data security, Luke made it clear that organisations cannot treat data in isolation.AI risk connects to identity security, device security, application security, cloud security, network controls and user behaviour. A strong data security strategy must sit within a wider cyber resilience approach. If an attacker compromises a user account with access to sensitive files, those files may still be exposed, even if the organisation has strong classification and encryption policies in place.
AI brings real opportunities for productivity, automation and innovation. Without the right controls, it can also increase the risk of data leakage, shadow AI, overexposure and compliance gaps.
Organisations that benefit most from AI are those that build strong foundations: robust identity controls, clear data classification, effective DLP, user education and continuous monitoring.CyberOne’s message is clear: AI adoption should be driven by confidence, not fear, and supported by the right security strategy.
For organisations unsure where to start, CyberOne can help assess current data and AI risks, identify gaps and build a practical roadmap for secure AI adoption.
Don't worry! The session is now available on demand. Watch the recording to learn practical strategies for protecting sensitive data, strengthening AI governance, and reducing data security risk in an AI-driven world.
Interested in improving your data security posture?
Get in touch to speak with one of our Microsoft security specialists and discover how CyberOne can help you secure your data with Microsoft Purview and a proactive, risk-based approach.