How Localiza reorganized work around generative AI
A new operating model
Localiza views generative AI not as a supplementary tool, but as a fundamental change in how the company operates. By organizing its collaboration with OpenAI around three strategic pillars—experimentation at scale, new customer interfaces, and AI-native products—the company has redesigned its internal processes to prioritize intelligence, context, and speed.
“We see this as a powerful tool that helps our teams do what they already do well. They live our culture, understand our customers’ needs, and know how to deliver great service.” —André Petenussi, CTO, Localiza
Scaling internal experimentation
Localiza treats AI adoption as a mechanism for organizational learning. Rather than limiting access, the company distributed ChatGPT licenses across all departments, including finance, legal, and operations. This approach has allowed non-technical staff—such as lawyers, analysts, and project managers—to build their own custom GPTs and prototypes. This shift has redefined the development lifecycle, allowing multidisciplinary teams to prototype solutions before they reach the formal development queue. The primary challenge, according to the company, is not technical, but the human-centric task of upskilling its 19 tribes and 130 squads.
Conversational customer interfaces
Recognizing that customers increasingly use natural language to express intent, Localiza has integrated its services directly into conversational platforms. In March 2026, the company launched an app integrated into ChatGPT, allowing users to explore rental and subscription options directly within the chat interface. This integration is treated as an experiment to keep the company aligned with evolving customer behavior and to gather data on how AI complements existing mobility services.
Liza: An AI-native conversational assistant
Liza is the company’s primary AI-native product, designed to handle complex, repetitive customer journeys. Unlike traditional chatbots, Liza is built on an architecture of intent and context, utilizing OpenAI models and specialized tools to manage the rental experience. The company emphasizes that success is measured by the quality of the customer experience rather than automation volume alone. By mapping customer journeys and establishing quality metrics prior to automation, Localiza ensures that human intervention remains available for high-value interactions.
Strategic results and lessons
Localiza’s implementation has yielded several key operational outcomes:
- Market Presence: Localiza was the first mobility platform in Brazil to integrate directly into ChatGPT and has since joined the platform's advertising pilot.
- Operational Scale: The Liza assistant now operates with over 60 specialized agents powered by reasoning models and tool orchestration.
- Organizational Shift: The company has successfully moved beyond IT-led innovation, with business units actively developing their own automations.
For leadership, the company highlights that the adoption bottleneck is reskilling, not the underlying technology. They recommend mapping and measuring customer journeys before attempting automation and treating AI presence in new channels as a long-term learning experiment rather than a short-term acquisition tactic.