Rolling out AI without a strong security strategy can expose businesses to escalating cyber threats. As AI adoption accelerates, leaders must balance innovation with risk management. Industry experts outline five essential security tactics organizations should implement to scale AI safely and responsibly.
First, share cybersecurity knowledge across teams. Barry Panayi, Group Chief Data Officer at Howden, emphasizes cross-functional collaboration between AI specialists and security professionals. AI security is not solely a technical issue, it requires shared accountability and communication across departments.
Second, return to foundational security principles. Nick Pearson, CIO at Ricoh Europe, stresses that AI security still depends on established best practices such as secure-by-design architecture, strong data governance, and standardized controls. AI should integrate into existing cybersecurity frameworks rather than operate outside them.
Third, treat AI as an assistive tool, not an autonomous decision-maker. Martin Hardy of Royal Mail highlights the importance of governance forums and visibility into how data flows into AI systems. Organizations must understand what data is used, how outputs are generated, and where oversight applies.
Fourth, recognize shared liability risks. Gartner’s John-David Lovelock warns that many AI vendor agreements may shift responsibility for safety onto end users. Businesses must review contracts carefully and assume accountability for AI-related security outcomes.
Fifth, embed AI directly into security workflows. Jeff Love, CTO at the Professional Rodeo Cowboys Association, uses AI tools to audit code, detect vulnerabilities, and improve logic before deployment. AI can enhance oversight by identifying risks humans may overlook.
主な原則は次のとおりです。
- Cross-team security collaboration
- Strong data governance foundations
- Clear accountability for AI risk
- Vendor contract scrutiny
- AI-driven code and logic review
The broader conclusion is clear: AI security requires structured governance, shared expertise, and integration into existing processes. Businesses that combine automation with disciplined oversight can innovate without compromising resilience.
ソース：
https://www.zdnet.com/article/5-ways-to-ensure-great-security-in-an-age-of-ai/
