I've been looking into Harvey/Legora's competitive moat for a paper I'm writing. The claim that 95% of what firms are paying for is the data security offered by Harvey/Legora's API buffer is one that I see often, but now that foundation-model providers (Anthropic, OpenAI, etc.) offer zero-retention enterprise tiers and configurable data-residency options that allow firm-side systems to mediate document access without exposing privileged material to the model provider, is there really still any meaningful security benefit offered by Harvey/Legora?
Comments
I've been looking into Harvey/Legora's competitive moat for a paper I'm writing. The claim that 95% of what firms are paying for is the data security offered by Harvey/Legora's API buffer is one that I see often, but now that foundation-model providers (Anthropic, OpenAI, etc.) offer zero-retention enterprise tiers and configurable data-residency options that allow firm-side systems to mediate document access without exposing privileged material to the model provider, is there really still any meaningful security benefit offered by Harvey/Legora?