Billionaire investor Mark Cuban argues that AI memory creates high switching costs for consumers. He suggests that users remain tied to specific chatbots because moving accumulated data between platforms is currently difficult compared to standard file formats like Excel or Word.
Cuban contends that switching costs will only decrease once a standardized export format is established. While some users argue that current models can process various file types, Cuban maintains that the manual effort required to transfer history remains a significant barrier to platform migration.
The industry is already seeing moves toward interoperability. In March, Anthropic enabled Claude users to import chat histories from competitors like ChatGPT. Additionally, the Model Context Protocol, introduced in 2024, allows AI models to connect to external data sources, potentially decoupling memory from individual chatbots.
Market competition is intensifying as companies vie for user loyalty. Anthropic and OpenAI released lower-cost models in September, shortly before Anthropic's planned IPO in November. Microsoft has also begun integrating its own models into Excel and Outlook, reducing friction for business users.
Despite technical solutions, some observers suggest that user habits are a primary factor in platform retention. While data remains portable through manual exports or by running models locally, the first company to simplify the migration process to a single click may secure a competitive advantage.
