Among the many conversations that defined Commercial UAV Expo's three days in Las Vegas, Thursday's keynote stood out for tackling a persistent industry challenge: why do so many drone programs get stuck in endless pilot phases, never quite making the jump to scaled operations?
‍

The "Breaking Silos, Building Skies" session on September 4 brought together Chris Fleming, CEO of Cyberhawk; Scott Lashmit, Aviation Manager at Cyberhawk; Michelle Duquette, founder of Duquette Consulting and former FAA official; and Nitin Gupta, founder and CEO of FlytBase. Their discussion provided answers, and a framework that could change how organizations approach autonomous operations.
‍

As Lee Corkhill, Group Event Director at Commercial UAV Expo, said in an earlier press release: "This keynote is about more than collaboration. It is about cutting through the noise and charting a smarter path forward for commercial drone operations."
‍

The Collaboration Paradox Nobody Talks About

Chris opened the session by addressing something many attendees could relate to: despite improving regulations and advancing technology, implementing drone programs at scale often feels more complex than it should. Organizations invest heavily in compliance, secure waivers, and demonstrate technical capabilities, yet somehow never quite establish routine operations.

‍

The panelists suggested this stems from treating collaboration as optional rather than essential to operations.

‍

Michelle's perspective proved particularly valuable, drawing from her experience inside the FAA. She pointed to the airline industry's evolution in the 1990s, when carriers moved from resisting information sharing to embracing collaborative decision-making as fundamental to efficient operations. She noted that the public safety community already demonstrates effective collaboration through tools like the Tactical Awareness Kit for real-time information sharing across agencies.

‍

The suggestion: commercial drone operators could adapt these proven models rather than developing entirely new approaches.

‍

A Framework for AI Integration

One of the session's most discussed topics was Nitin's approach to AI adoption, which addressed concerns many organizations have about autonomous systems. He presented a two-category framework that distinguishes between different types of AI applications.

• High-risk AI involves direct flight control such as detect-and-avoid systems, autonomous navigation, and flight path optimization. Here, errors could cause crashes or airspace violations, making adoption naturally slower and regulator-driven.
  ‍
• Low-risk AI focuses on processing data streams that drones generate, including anomaly detection, image analysis, predictive maintenance insights, and operational optimization. The key difference: errors in this category don't cause accidents, making it a safer entry point for AI integration.

This distinction offers organizations a path to begin AI adoption immediately by running systems in shadow mode alongside human analysts. The approach builds operational trust and generates audit trails while avoiding safety risks.

‍

Building on the AI Collaboration Concept

Nitin's emphasis on AI as a collaborative partner suggests a broader progression that could benefit the industry. His framework points toward a potential maturity model: Compliance → Collaboration → Autonomy.

‍

Currently, most organizations focus heavily on compliance—securing approvals and meeting regulatory requirements. But Nitin's approach of treating AI as a silent partner points toward something more: using these collaborative relationships to build the operational routines and trust networks necessary for eventual autonomous capabilities.

‍

This progression also aligns with what Scott described in his BVLOS experience, where collaborative partnerships with the FAA built on transparent safety cases and data sharing have proven more effective than purely compliance-driven approaches.

‍

Multiple Stakeholder Coordination

The discussion highlighted how today's drone ecosystem requires coordination across five distinct actors: end users, service providers, technology providers, regulators, and AI systems. Each brings different priorities and constraints, but successful operations require alignment across all five.

‍

This multi-stakeholder reality explains why isolated approaches struggle. Technology providers build solutions without understanding operational constraints, operators pursue certifications without considering end-user needs, and regulators develop frameworks without industry input. All of these contribute to the fragmentation that keeps programs in pilot phases.

‍

Three Immediate Takeaways

As part of Commercial UAV Expo's broader "Drones in Action: Current Realities and Future Frontiers" conference program, the session marked a definitive shift in industry thinking: from viewing AI as a future threat to understanding it as an immediate collaborative tool. The keynote delivered specific next steps for industry participants looking to build effective partnerships and align project outcomes:

• Start low-risk AI integration immediately. Data processing automation provides safe entry points for building operational experience and regulatory confidence without waiting for autonomous flight approvals.
  
• Adopt proven collaboration models. The frameworks aren't theoretical—they're working in public safety and specialized industrial applications. Commercial operators can adapt these models rather than developing new approaches.
  
• Build operational routines now. Systematic, documented operations at any scale create competitive positioning that compounds over time.

Whether this collaboration and AI framework gains broader adoption will depend on how successfully early implementers demonstrate its practical value. But for an industry that continues to grapple with scaling challenges, the Thursday keynote provided concrete strategies for moving beyond compliance-focused approaches toward more collaborative, AI-enabled operational models.