Archer Launches Zee, An AI Foundation Model For Aviation

Archer Aviation announced Wednesday the development of Zee, a new aviation-specific AI foundation model designed to serve as a unified intelligence platform for the aerospace sector. The technology integrates multiple disparate aviation data streams, including Automatic Dependent Surveillance-Broadcast (ADS-B) signals, air traffic control (ATC) communications, aeronautical maps, charts, aircraft-state metrics, terrain data, and real-world weather variables.

Developed as a primary component of Archer’s broader physical AI strategy, the foundation model is engineered to operate both as a server-hosted solution and locally on-device without requiring external network connectivity. Company officials stated this dual-operational capability is critical for deployment across diverse aviation environments, including electric air taxis, unmanned aerial vehicles (UAVs), commercial airliners, and air traffic management centers.

“We are building an intelligence layer for the entire aviation system with Zee. The company that owns the data and the foundation model will help lead the aviation industry into the next era of flight,” said Adam Goldstein, founder and CEO, Archer.

The model’s training utilized real-world operational data gathered through Archer’s proprietary data pipeline and an established global infrastructure network featuring more than 6,000 ADS-B receivers. According to company data, more than 45,000 flights traverse US airspace on a typical day, generating a continuous volume of radio transmissions, navigation inputs, and mechanical status updates that pilots and air traffic controllers must currently synthesize manually in real time.

Archer designed Zee to process these diverse variables as a single, interconnected system to achieve lower latency and higher operational performance under constrained conditions.

The development program is managed by an internal AI division consisting of nearly 100 researchers and engineers. The team is directed by Mario Srouji, vice president of AI products, Archer, who joined the firm following a previous tenure at Apple. The division is advised by Professor Ruslan Salakhutdinov, the former vice president of AI research at Meta and former director of AI research at Apple, alongside researchers drawn from Silicon Valley technology companies and academic laboratories.

“Aviation is having its GPT moment,” Srouji said. “Our current national air system is built on legacy technology, and is ripe for AI innovation. Archer’s world-class team, and industry-leading Zee foundation model positions us to help lead the industry transformation; unlocking new levels of safety, efficiency, and scale for America’s airspace.”

The deployment of the model comes amid broader infrastructure modernization initiatives. The US Department of Transportation has allocated approximately US$20 billion toward updating the national airspace system, an investment that industry analysts and participating companies cite as validation for upgrading legacy operational infrastructure.

Archer confirmed it is currently engaged in preliminary discussions with government agencies, commercial airlines, and international aerospace partners to deploy the foundation model through initial pilot programs. The company has designated airline operations, automated airspace management, and copilot flight assistance software as the primary near-term applications for the platform, with the stated objective of lowering operational latency and mitigating safety inefficiencies.