Harvey II Launches Innovative Memory-Based Legal AI Tools
The legal industry is witnessing a new era of personalized artificial intelligence with the launch of Harvey II, a next-generation legal AI platform. The standout feature of Harvey II is its enhanced memory-based legal AI capability, which enables lawyers and legal teams to streamline their workflows and deliver more consistent outcomes across cases and matters.
Personalization at the Core: How Memory Works
Harvey II’s new Memory feature is designed to learn and retain an individual lawyer’s drafting style, structural preferences, citation habits, and level of detail over time. This means that every interaction with the AI becomes more tailored and efficient, reducing the need for repeated instructions with each new case. By remembering these preferences, the platform ensures lawyers can work in their preferred style across the entire Harvey system, as well as in widely used applications like Microsoft Word and Outlook.
Unlike many AI platforms, Harvey II allows users to review, modify, or turn off what the system remembers. Importantly, information stored in Memory is not used to train external models, keeping sensitive data secure and private—a key requirement in the legal sector.
Spaces: Organizing Around Matters and Projects
Alongside its memory-based legal AI, Harvey II introduces a feature called Spaces, which is designed to centralize all work related to a particular matter or project. Spaces serve as collaborative hubs where documents, tasks, permissions, and work history are stored, ensuring both lawyers and AI agents operate with access to the same context. This approach encourages seamless collaboration and reduces errors that can arise from fragmented information.
Tasks within a Space can be assigned to either a human lawyer or a software agent. Permissions and ethical walls can be synced directly from a firm’s existing systems, ensuring compliance with internal access controls and client confidentiality. Agents can move between Spaces without transporting client data, reflecting the legal industry’s strict standards around privacy and data separation.
Developed in Partnership with Leading Firms
The development of Harvey II’s memory-based legal AI features involved close collaboration with law firms and in-house legal teams across the US, UK, Europe, the Middle East, and Asia-Pacific. Among the contributors was Mallesons, which played an active role in shaping the Memory feature. Michelle Mahoney, Chief Innovation Officer at Mallesons, emphasized the advantage of capturing and applying individual preferences and drafting styles, stating that personalization makes AI feel like an authentic extension of the legal team.
This collaborative approach ensured that Harvey II was built with real-world legal workflows in mind, including robust support for ethical walls, confidentiality, and granular access controls.
A Three-Phased Rollout for Memory Features
Harvey II’s memory-based legal AI functionality is being introduced in three distinct phases. The first phase, already in early access, focuses on Personal Memory, capturing preferences related to drafting, structure, tone, and citations for individual lawyers. The second phase will extend memory capabilities across shared matters or Spaces, enabling teams to benefit from collective knowledge and context. The final phase will allow organizations to implement broader styles of working while maintaining strict boundaries between users, matters, and clients.
This phased approach reflects a broader industry challenge: delivering meaningful AI-powered personalization while preserving the confidentiality and information barriers essential to legal practice.
Shifting the Paradigm of Legal AI
Harvey II’s update signals a shift from one-off, prompt-based AI tools to platforms designed around ongoing work and persistent context. By organizing AI around matters and projects, the system facilitates a more active role in routine legal tasks such as drafting, matter management, and collaboration. This represents a significant step forward for memory-based legal AI, positioning Harvey II as a leader in the move toward smarter, context-aware legal technology.
Conclusion: The Future of Memory-Based Legal AI
For law firms and legal departments, Harvey II’s memory-based legal AI offers the promise of increased efficiency, better compliance, and more personalized workflows. As the platform continues to evolve, it is likely to set new standards for how AI integrates with legal practice, making legal technology a more seamless and integral part of daily professional life.
This article is inspired by content from Original Source. It has been rephrased for originality. Images are credited to the original source.