Panel discussion at Flower AI Summit 2026

At Flower AI Summit 2026, Prof. Shiqiang Wang joined a panel discussion on “Accelerating the Adoption of FedAI”, alongside experts from healthcare, security, and engineering practice. The discussion focused on a central message: the technical foundations for federated AI are increasingly mature, but broad adoption now depends on governance design, legal interoperability, and clearer incentives for organizations and practitioners.

Across the panel, several recurring barriers emerged. Legal and regulatory complexity remains a major bottleneck, especially when collaboration spans jurisdictions and sectors with different compliance regimes. At the same time, organizational incentives are often misaligned: many teams still frame data as a competitive moat rather than a shared infrastructure for prevention, safety, and system-level efficiency.

A second major theme was implementation realism. Beyond model quality, participants emphasized staffing constraints, data engineering readiness, and the challenge of scaling federated systems under heterogeneous infrastructure conditions. In healthcare and finance, practical deployment requires not only privacy-preserving learning, but also measurable links to operational and funding metrics that decision-makers already use.

Shiqiang highlighted that the scope of federated AI is expanding from model training to a broader federated AI toolchain, including data processing, code generation workflows, and collaborative knowledge artifacts. He also emphasized human productivity as a key adoption driver: when AI systems can reliably automate substantial portions of everyday work under clear governance, willingness to participate in federated collaboration increases significantly.

The panel converged on a pragmatic path forward: begin with focused, trust-based federations, demonstrate measurable value in constrained settings, and then scale through stronger standards, clearer governance models, and interoperable technical protocols.