Cloudera, the only company bringing AI to data anywhere, is warning that without female representation in senior decision making roles, the accelerated push to deploy agentic AI risks embedding systemic gender bias into autonomous systems.
According to IDC research, by 2027 half of enterprises will rely on AI agents to reshape how humans and machines work together, making 2026 a pivotal year for moving agentic AI into production. As these systems take on business-critical decisions, including workforce planning, pay recommendations, and operational prioritisation, Cloudera warns that any bias, propagated by poor data and a lack of diversity in building AI, will be amplified. These gaps in leadership and development could lead to discrimination in hiring, salary packages and company-wide operations.
This warning follows recent Cloudera research, with UK female IT leaders highlighting growing fears over who is shaping AI decision-making today:
- 68% are concerned about the lack of women in senior AI roles
- 56% believe this gap will lead directly to biased AI outputs
- 57% say AI is already inherently biased because AI company leadership remains predominantly male
“Without the correct oversight and guardrails, agentic AI dramatically raises the stakes when it comes to risk,” says Manasi Vartak, Chief AI Architect at Cloudera. “When AI systems can act on their own, any fragilities – whether in the data feeding them, how they’re designed, or who designed them – is massively amplified. A small crack in development can become a serious fault line once that system is operating autonomously, making decisions that affect people, businesses, and society.”
Cloudera stresses that the issue goes beyond representation alone, and that effective agentic AI governance depends on having the right mix of voices and roles at the leadership table – people who build, challenge, test, and oversee AI systems. This must also be supported by strong data foundations that ensure agents are trained on accurate, trusted, and well-governed information. Cloudera warns that without diverse perspectives at the decision-making table, organisations are far more likely to miss risk, overlook bias, and deploy systems that fail under real-world conditions.
“AI must be representative of the people and communities it is designed to serve,” Vartak adds. “That only happens when diverse perspectives are shaping decisions from the start. Leadership teams need to be empowered to question results, stress-test assumptions, and challenge the data going in. Diversity of thought, combined with high-quality data, helps ensure AI scales responsibly and delivers reliable outcomes.”
As organisations move into the agentic era, Cloudera offers the following advice to those deploying AI:
- Review AI leadership structures to ensure women are represented in senior AI decision-making and strategic roles
- Build leadership teams with a balance of technical, governance, and challenge-oriented roles
- Ensure AI systems are provided with high-quality, trusted, and well-governed data to avoid ‘bad data in, bad decisions out’
- Treat diversity in AI leadership as a core component of AI risk management, not a secondary initiative
