Nuix launches AI agent for forensic investigations

Nuix has launched the Nuix Neo AI Agent, adding agentic AI to its Nuix Neo investigation platform.

The software is aimed at legal, regulatory and criminal investigations, where users need to trace how a result was reached. It answers plain-language questions, links each answer to source material and records the steps taken to produce a result.

The launch expands Nuix's use of artificial intelligence in a market where auditability and evidence handling remain central concerns. Its software is used to analyse large volumes of unstructured data, including emails, chats, documents and images, in matters involving compliance monitoring, legal review and law enforcement.

Audit trail

A central feature of the new tool is a built-in record of its activity. The AI Agent logs the documents it reads, the queries it runs, the connections it follows and the outputs it returns.

According to Nuix, that record is created at the point of execution rather than assembled later. Users can export the full conversation, including the reasoning path and findings, for peer review, court scrutiny or board reporting.

In practice, the system is designed to let investigators ask questions in natural language while retaining a documented path back to the underlying evidence. Every answer is linked to its source documents so claims can be checked against the original material.

"For more than two decades, organisations have trusted Nuix with their high-stakes matters. Agentic AI should raise that bar, so we built the Nuix Neo AI Agent to show its work. Trace every answer to its source. Review every step. Rely on forensic-grade data processed by the Nuix Neo platform," said John Ruthven, chief executive officer of Nuix.

How it works

The AI Agent is built on the Nuix Neo Gen AI Framework, which connects supported large language models with data processed through the Nuix Neo platform. The agent can work across entire Nuix Neo case inventories and use more than 20 tools available in the platform.

It also integrates with the Nuix knowledge graph and can generate graph queries automatically, reducing the need for users to write them manually. The system is intended to identify and follow links between people, accounts, communications and transactions across datasets.

Nuix says its underlying processing engine reads data at the binary level across more than 1,000 file types, including corrupted, encrypted and legacy files. It also says processing is deterministic, meaning the same input produces the same output each time.

Governance focus

Nuix is positioning the release around governance and oversight rather than full automation. It says the framework was built to address concerns about confidence, auditability and control that have slowed wider use of generative AI in regulated industries.

According to the company, AI-generated content is clearly marked and outputs are grounded in source material that can be verified. The system is intended to support the judgement of legal, compliance and investigative professionals rather than replace it.

Administrative controls are another part of the offer. Organisations can set spending caps, monitor token use, define prompt guardrails and restrict access by role.

The product can be deployed in cloud or on-premises environments, allowing customers to align the setup with internal security and data sovereignty requirements. Role-based access controls are enforced throughout the system, according to Nuix.

Market context

The release comes as software groups seek to add generative and agentic AI features to tools used in investigations, e-discovery and compliance work. In those fields, customers often need to show not only what a system found but also how it reached that result.

That requirement has become more important as organisations test AI tools in matters that may face regulatory review, litigation or criminal proceedings. Systems that cannot provide a clear chain between source evidence and output may face limits on their use in those settings.

Nuix has long focused on organisations handling sensitive data and matters under scrutiny, including government agencies, financial institutions, law firms and regulators. Its software has been used in investigations ranging from major cross-border leaks to day-to-day compliance and review work.

The new product adds AI-driven query and analysis functions to a platform where repeatability, documentation and access control remain key buying factors.