Responsible AI Starts With The C-Suite

Forbes

30-06-2025

Shane Buckley is President and Chief Executive Officer of Gigamon, a leader in deep observability.
AI is at the top of every board agenda today. With global AI investment expected to surpass $200 billion in 2025 and $750 billion by 2028, the central conversation has shifted to balancing innovation with responsible data use.
As AI matures, so do its risks—particularly those related to privacy, security and ethics. To scale responsibly, business leaders must find equilibrium between aggressive AI adoption and intentional governance.
As I emphasized in my last Forbes Technology Council article, AI budgeting must begin with a security-first mindset. Today, that mindset is no longer optional. It's a strategic imperative for the C-suite—one that sets the foundation for sustainable, scalable success.
Data Risk Is The Core Challenge
At the center of AI-related risk lies a fundamental uncertainty: how large language models (LLMs) process, retain and expose sensitive data. These models often contradict Zero Trust principles by opening broader access to networks and information. When confidential business data is entered into third-party AI tools, it may be stored in jurisdictions with incompatible compliance or privacy laws.
Employee and third-party misuse—intentional or not—can expose organizations to data leakage, regulatory risk or public breach.
According to my company Gigamon's 2025 Hybrid Cloud Security Survey, which included over 1,000 IT and security leaders, visibility into data in motion is now a top business priority. More than half (54%) of respondents expressed reluctance to use AI in public cloud environments due to intellectual property risks, while seven in ten are considering moving data from public to private clouds.
The Internal Threat Is Often Overlooked
While headlines focus on deepfakes and AI-enabled phishing attacks, a quieter threat looms within: employees unknowingly inputting sensitive data into unsecured AI tools. Even well-intentioned teams can become the weakest link if the organization lacks appropriate controls.
A Growing Dark Market For Malicious AI
On the dark web, malicious AI tools—black-hat versions of ChatGPT—are enabling adversaries to launch more frequent and sophisticated attacks. Our survey found that 58% of leaders observed an increase in AI-powered ransomware. In 2024 alone, there was an 11% increase global spike in reported ransomware attacks, with over 5,400 attacks logged.
The convergence of an expanding threat surface and rapidly advancing attacker capabilities makes a reactive cybersecurity strategy untenable.
Accountability Must Extend To Vendors
As AI use becomes embedded in third-party systems, vendors represent a growing risk surface. Increasingly, companies are requiring detailed disclosures on how their partners use AI. Transparency, accountability and aligned standards across vendors are critical.
If even one supplier is compromised, AI-powered malware can cascade through interconnected systems and impact entire supply chains. Business leaders must extend security-first thinking to external partnerships and vendor ecosystems.
When AI is accessible without oversight, organizations risk losing control over their data footprint. But bans aren't the answer—these only encourage unmonitored "shadow AI" use. Instead, responsible enablement must prevail. That includes educating employees, enforcing clear policies and building visibility across enterprise AI stack.
Boards Must Lead Governance
AI governance is no longer the sole domain of IT. It's a board-level issue.
Forward-looking organizations are forming AI governance committees that include the CEO, CISO, CRO and General Counsel. These cross-functional teams are tasked not only with risk oversight, but also with defining the organization's risk appetite, monitoring AI use and maintaining compliance across jurisdictions.
True governance is more than policy—it's cultural. It ensures AI is used safely but applied in ways that benefit both people and the business.
Ethical Risks Carry Legal Consequences
Security isn't the only concern. AI systems can carry ethical reputational risks—from bias to misinformation.
Bias in AI can lead to discriminatory results. Hallucinations, when models generate convincing but false information, can mislead decision-making and create legal exposure. Organizations can mitigate these risks through measures like pseudonymization—removing personally identifiable information (PII) before inputting the data into AI systems.
Even simple steps—such as stripping customer or vendor names—can improve privacy protection and reduce harmful outcomes.
Prioritize Responsible Innovation
AI has transformative potential, but only for organizations that wield it with care.
C-suite leaders must guide their organizations through bold innovation while safeguarding core values—people, data and trust. Those who take a security-first, governance-led approach today will shape the AI-powered businesses of tomorrow.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?