Navigating the Future in Azure: AI, Zero Trust, and Data Security

Navigating the Future in Azure: AI, Zero Trust, and Data Security

On 24 July 2025, the Executive Leaders Network, in partnership with Atech and iomart, hosted a webinar exploring how organisations can navigate the rapidly changing Azure landscape. The session, titled Navigating the Future in Azure, brought together cloud specialists Danny Nicholson and John McKelley to share their expertise on AI adoption, zero trust security, and data governance.

The discussion, moderated by Peter, highlighted practical steps for aligning cloud innovation with business needs, while mitigating the evolving risks of a more connected, AI-driven world.

Speakers:

Danny Nicholson – Public Cloud Specialist Lead, Atech
With over 15 years’ experience in Microsoft public cloud solutions, Danny has worked across industries helping organisations adopt Azure effectively. His focus is on solving real-world challenges in cloud adoption, balancing innovation with operational stability.

John McKelley – Cloud Specialist, Atech (part of iomart group)
John specialises in modern workplace and Microsoft cloud journeys. He advises organisations on rightsizing their Microsoft services, ensuring they gain maximum value while maintaining compliance and security.

Key Topics Covered

1. The Evolving Azure Landscape

Azure has grown from around 65 services less than a decade ago to over 250 today, with more than 100,000 SKU options. Originally seen as a space for testing and innovation, Azure is now core infrastructure for most organisations.

Market pressures and customer expectations are driving rapid migration from traditional data centres into public cloud. The challenge is keeping pace with this shift — particularly in security — without losing operational control.

2. A New Threat Terrain

Cloud adoption has expanded the attack surface, especially in hybrid and multi-cloud environments. Key risk areas include:

• Shadow IT – Employees using unapproved tools and platforms.
• Misconfigurations – Still among the top five causes of successful cyberattacks.
• AI-enabled threats – Malicious actors using AI for tasks such as port scanning, phishing, and exploiting vulnerabilities at scale.

A notable example shared was the Chevrolet AI chatbot breach (Dec 2023), where a poorly configured AI assistant was manipulated via prompt injection to offer a $76,000 car for $1.

3. Zero Trust as a Mindset

Zero trust is often mistaken for a product or architecture, but as Danny explained, it is primarily a mindset:

“Verify everything, trust nothing.”

This approach focuses on continuous verification — of identity, devices, applications, and data — with micro segmentation to prevent lateral movement by attackers. Importantly, zero trust must also extend to human behaviours, embedding a culture of security awareness across the organisation.

4. AI in Security — AI vs AI

AI is now both a security challenge and a defence mechanism. Microsoft, working with OpenAI, has embedded AI capabilities across its Security Co-pilot platform to detect, analyse, and respond to threats in real time.

Capabilities include:

• Anomaly detection inside and outside the network perimeter.
• Identifying compromised accounts or systems mid-attack.
• Analysing millions of signals per second to counter AI-driven attacks.

The takeaway: security tools should have AI capabilities built in — not as add-ons — to keep pace with modern threats.

5. Protecting Data in an AI-Driven World

Data remains the “crown jewels” of any organisation. While encryption at rest and in transit are standard, data in use is often overlooked — and is at its most vulnerable when stored temporarily in clear text in system memory.

Strong data strategies must include:

• Classification and labelling.
• Access controls and auditing.
• Secure handling of both structured and unstructured data.

Organisations considering tools like Microsoft Copilot should begin with a data readiness assessment to ensure information is properly classified and protected before scaling AI use.

Practical Starting Points

For those starting their Azure AI journey, Danny advised focusing on small, high-value proof-of-concept projects. One example was automating invoice validation using Microsoft’s Document Intelligence service — a simple, non-disruptive step that can free significant staff time while maintaining human oversight.

John emphasised the importance of building a measurable business case and exploring Microsoft funding opportunities, which can be unlocked by working with accredited partners such as Atech.

Q&A: Expert Answers for Common Azure & AI Security Challenges

Q: I’m just starting with AI in Azure. Where should I begin?
A: Start with a clear business objective. Identify a single, measurable process that AI could improve without high disruption. For example, automate repetitive tasks such as invoice validation using Azure Document Intelligence. Keep it small, prove value quickly, and then scale with guidance from a partner who can ensure security and compliance from the outset.

Q: How can my organisation fund an Azure AI project without committing huge upfront costs?
A: Microsoft offers funding programmes for qualifying projects, especially if they demonstrate measurable business impact. Working with a Microsoft-accredited partner like Atech increases your chances of securing this support, as they can navigate the qualification process and align it with your business case.

Q: What’s the fastest way to improve Azure security without overhauling everything?
A: Adopt zero trust in stages. Begin with one pillar — for example, securing and continuously verifying device access. Combine this with user awareness training so the mindset becomes part of everyday operations.

Q: Is Copilot safe to roll out to my whole organisation immediately?
A: No — start with a readiness assessment. Check that your data is classified, labelled, and access-controlled before scaling. Without these guardrails, you risk exposing sensitive information unintentionally through AI queries.

Q: What’s the most overlooked data security risk in the cloud?
A: Data in use — information stored in memory during active processing. It’s often left unprotected and may be in clear text, making it vulnerable to advanced attacks. Ensure your security tools protect data at rest, in transit, and in use.

Key Takeaways

1. AI is transforming Azure — but security strategies must evolve in parallel.
2. Zero trust is a mindset, not a product — start with small, high-impact measures.
3. Data governance is non-negotiable — especially before deploying generative AI tools.
4. Start small — identify a targeted process where AI can deliver measurable value.
5. Leverage funding and expert support to accelerate adoption while managing risk.

Watch the Webinar On-Demand

To explore these topics in full and hear real-world examples from Danny and John, watch the full recording here:

Watch On-Demand

This session reinforced that Azure adoption, AI integration, and zero trust security are no longer “future” concerns — they are today’s business imperatives. Organisations that align technology adoption with clear objectives, rigorous security, and robust data governance will be best positioned to compete in the next phase of cloud evolution.

#AzureCloud #MicrosoftAzure #CloudSecurity #ZeroTrust #AIAzure #DataGovernance #CyberSecurity #CloudTransformation #DigitalWorkplace #AITools #CopilotForBusiness
#AIvsAI #SecureCloud #CloudMigration #AIProductivity