Augmetec vs Nuix Neo: how they compare in 2026

AI functions named on the platform page, as updated in November 2025, are identifying relevant evidence and building fact patterns against a Line of Inquiry using what the vendor calls accurate AI algorithms, document fraud detection, automatic identification of witnesses and preparation of interview questions, chronology building, and AI driven triage, scoping and categorization of cases. Underneath sits a full case management system that does a great deal on its own: a central data lake, evidence management, templates, customizable workflows, dashboards, collaboration, single sign on, enterprise integrations and automated evidence bundles. Without the models, an investigations team still has a working command center. The master services agreement never mentions artificial intelligence in its operative terms.

Nuix Neo's AI classifies and contextualizes data, applies language models described as trained for the customer's use case to cut review volume, and drives automated redaction. Without it, the Nuix Engine still processes more than a thousand file types into searchable form. Early case assessment, review in Nuix Neo Discover and investigation workflows also remain, which is the product Nuix has sold for years. The Solutions Terms of July 2026 say Nuix will identify in its documentation the features that materially rely on AI systems, and the license is sold by data volume processed rather than by AI use.

The platform page says a team can answer questions defensibly, that investigations are run accurately and defensibly, and that evidence identification uses accurate AI algorithms, alongside industry leading document fraud detection. None of those claims carries a figure, test set, evaluation, error rate or statement of failure modes. Grounding is not described: nothing says whether a finding surfaced against a Line of Inquiry links back to the document and page it came from. Clause 2.5 of the agreement has the customer acknowledge that the software may contain bugs, errors, defects and design flaws and may produce unexpected results.

Nuix says its language models improve accuracy and reduce the data to review. Its telecommunications page claims that up to 95 percent of irrelevant data is eliminated early, with no test set, method or recall figure. The Solutions Terms of July 2026 call AI output probabilistic and warn that it may contain errors, omissions or bias. They disclaim any warranty on its accuracy, completeness or reliability and make the customer responsible for verifying it. The product classifies and culls documents rather than citing law, and no measured error rate for its classification is published.

The platform page says case scoping is semi automated, smart actions suggest team actions, and witnesses are identified and interview questions prepared automatically in real time. Nothing says which outputs need review before use, what a reviewer sees, when the system acts without a person, what happens when a fraud detection flag is wrong, or whether generated material is marked as generated. No abstention behavior or confidence signal is described. The agreement assigns the consequence without describing a mechanism: under clause 10.2 the customer bears sole responsibility for results and conclusions drawn from the software.

The models work alongside reviewing people under a written commitment, with real review surfaces, but the full control structure is not published. The review product is described as responsive review with workflows enabled by AI that support faster and more precise human review. A named customer describes running its own AI within its own environment under its own compliance controls. Automated workflow templates run culling and redaction that a reviewer then works from. Nothing published sets the threshold at which a document is excluded from review, says what runs without a person, or states a route back after a misclassification beyond ordinary review.

No customer is named on the site, no logo wall appears, no case study exists, and no deployment is described or dated. The single customer voice is one unattributed sentence on the About page, carrying no name, role or organization, saying that LEIAA is clearly built by practitioners for practitioners. Two figures are published on the platform page and neither is attributed or explained: administrative effort alleviated by up to 45 percent, and the platform being up to 80 percent cheaper than other platforms, the second a competitive pricing claim rather than an outcome, naming no comparator. The founding team's own experience is set out in detail, including more than 150 investigations conducted by the chief executive before founding the company, a credential rather than deployment evidence.

Donn Hoffman, Chief Privacy Officer and Deputy District Attorney at the Los Angeles County District Attorney's office, is quoted on the Discover page for installation on premises. He speaks about running the office's own AI within its own environment, with no measured outcome. IAV GmbH is named on the same page as an ediscovery services customer. The data reduction figure of up to 95 percent on the telecommunications page carries no customer name or date. No named customer is joined to a figure.

Clause 8.3 of the master services agreement is the core: the customer keeps full title to Customer Content, and without consent Augmetec will not access, use or disclose it except as reasonably needed to support use of the software, answer support requests, or for other purposes the customer authorizes in writing. That is a closed list of purposes, not an open license. Clause 8.4 allows collection of usage and performance data only if it cannot identify the customer or its customers and is not confidential or privileged in nature. Clause 6 makes Customer Content the customer's confidential information for five years, and the addendum puts everyone processing the data under a strict duty of confidentiality. Each customer has a dedicated instance, and Annex II says segregation procedures keep other customers out. Nothing describes how privileged material is handled inside the platform, no walls between matters within a customer are described, and no model provider is named.

Customer data stays the customer's under Nuix's Solutions Terms of July 2026, which license Nuix to process it only to provide the solutions. They state that Nuix will not use customer data to train any AI systems, treat customer data as confidential information, and require prompt written notice of any compelled disclosure where the law permits. Installed on premises, the software keeps matter data in the customer's own environment, and the bring your own AI option keeps third party models there too. For SaaS, each customer's databases and file shares are logically separated and encrypted with AES 256. Privilege and work product are not addressed as such, and no separation by matter inside the review product is described.

No statement that Augmetec is not a law firm, no disclaimer that output is not legal advice, nothing on the professional judgment needed before a finding is acted on, no jurisdiction limit, and nothing on supervision or competence is published. The platform, about and security pages and the master services agreement contain none of these. The nearest provision is clause 10.2, under which the customer bears sole responsibility for results and conclusions, which allocates liability rather than saying what the product is not. The platform builds fact patterns, identifies witnesses, prepares interview questions and produces investigation reports in matters of fraud, bribery and misconduct, where a conclusion affects named people. Augmetec also publishes website terms of use.

AI output is probabilistic and may contain errors, omissions or bias, according to Nuix's Solutions Terms of July 2026. The terms say AI systems are no substitute for legal or other professional advice or for review by a trained professional, and the customer must verify output before relying on it. Buyers span law firms, corporate legal teams, law enforcement, financial regulators and telecommunications companies. Nothing describes how the product supports a supervising lawyer's duties, and no jurisdiction limit is named beyond export control and the territory clause.

No responsible AI page, principles statement, accountable owner for model behavior, evaluation regime before release, AI management system or certification such as ISO 42001 is published. Nothing addresses uneven output or describes how the system's determinations about named people are tested or governed, though it identifies which people are witnesses, flags documents as fraudulent and builds fact patterns about alleged misconduct. The governance material that exists is about security: an information governance committee and ISMS group, a person overseeing information security and compliance, and certified security staff. The chief security officer holds a doctorate in machine learning, a credential rather than a governance framework.

Nuix markets its review product as powered by ethical AI and describes language models trained to the customer's use case. Its Solutions Terms of July 2026 commit Nuix to governance processes designed to identify, assess and mitigate material bias in its proprietary AI output. Customers get a written route to report systemic bias, and Nuix commits to comply with AI laws that apply to it as a provider. No governance framework, ISO 42001 or equivalent, named owner, testing before release or finding about uneven classification across document types or languages is published.

Annex II of the published addendum runs to several pages: AES-256 at storage system and device level with separately keyed backups, TLS 1.2 or higher in transit, VPN and multi factor authentication, access control lists on a need to know basis with differentiated profiles, automatic account locking and idle logout, network separation through a DMZ, a web application firewall, logging of access, changes, entries and deletions, intrusion detection, disaster recovery and continuity planning, physical security of data centers, an ISMS group and a named security function, internal and external audits, penetration testing with remediation, and a responsible disclosure program. Incident practice is contractual: clause 2.13 commits to telling the customer of a security incident without undue delay and within 72 hours, and to cooperating with the customer's own reporting duties. Clause 2.14 covers deletion: destroy or return on written instruction at termination. No retention period is stated; data is kept as required to perform the agreement.

A personal data breach gets notice without undue delay under the data processing addendum of June 2026, with information and help in mitigating it. Nuix deletes or returns personal data at the end of the agreement, or at any time on written request, unless the law requires it to keep some. A SaaS customer has 10 days after its subscription ends to download its data under the Solutions Terms of July 2026. Nuix then deletes it, and data is unrecoverable from backups 90 days after the end. The Security Requirements of June 2026 set encryption, logical separation of each customer's data and an incident response program. On premises, the customer holds and backs up its own data. No subprocessor list is published.

Clause 9.2 commits Augmetec to indemnify the customer against direct damages from third party intellectual property claims connected to use of the software, with remedies at 9.3: securing continued rights, or replacement or modification on stated equivalence conditions. Clause 10.3 keeps liability for death or personal injury, fraud and willful misconduct outside all caps. Against that, clause 10.1 provides the software as is, with the customer assuming the liability. Clause 10.4 caps liability at the lesser of one year's subscription fees or twenty thousand US dollars, and the IP indemnity at the lesser of two years' fees or fifty thousand dollars, so the twenty thousand dollar ceiling applies to everything except the IP indemnity and the 10.3 carve outs. The exclusions in 10.4 expressly cover loss of privilege and loss of confidentiality. Nothing warrants the accuracy of any output.

Third party claims that the customer's use of a solution infringes intellectual property are defended and any losses awarded or settled are paid under the Solutions Terms of July 2026. Exclusions cover unsupported versions, combinations and unauthorized use. Nuix warrants that software is free from defects for 90 days from the license key and that a SaaS offering is free from defects through the subscription, with repair as the remedy. Each party's liability is capped at the fees for the agreement year. A super cap of twice that applies to breaches of the security requirements or the data processing addendum that compromise customer data. Fraud, wilful default, death or injury from negligence and indemnified claims sit outside the caps. Nuix maintains professional liability and general liability insurance. AI output carries no warranty of accuracy, completeness or reliability.

Core enterprise integrations, single sign on and secure integrations with existing platforms, document management systems, and Microsoft and Google are promised on the platform page. That names the productivity and identity tools, and document management only as a category. Clause 4 of the agreement covers third party services the customer connects and single sign on, and says connecting a service authorizes Augmetec to access and store information from it. Nothing says what data moves between LEIAA and any outside system, in which direction, on what trigger, or what a firm must configure, and no API or developer documentation is published. The addendum mentions material from Microsoft Teams, Slack, WhatsApp, Telegram, Amazon Chime and Google Meet as categories of personal data that may be uploaded, not as connectors. Augmetec also runs a knowledge base on a third party support platform.

The Engine ingests more than a thousand file types and connects to data sources, which is the core of the product. Nuix's Solutions Terms cover customer integrations and the use of Nuix's APIs, and the Neo Legal page describes orchestrating business applications across the enterprise. No document, matter or review system connection is described with what moves, in which direction or what a customer must configure.

The security page says, under the heading your LEIAA, your environment, that each customer's instance runs on a separate dedicated environment available only to that customer, on a virtual private cloud built for data isolation. Clause 10.1 of the agreement repeats that each customer's instance is held on separate dedicated infrastructure, and the software is reached at a customer specific subdomain of leiaa.com. Annex II adds that customer data sits on its own instance with segregation procedures preventing access by other customers, and names Google Cloud Platform as the host. No region is named. The virtual private cloud is described as a global network across multiple regions, which describes reach, not residency. No region choice is offered and no residency commitment appears in the agreement, though the addendum includes transfer mechanisms through the EU standard contractual clauses and the UK addendum.

On the customer's premises or as a SaaS offering, Nuix Neo Discover serves organizations with data sovereignty and chain of custody requirements. On premises, the customer installs one production copy in its own environment and keeps complete control over its data, so residency is wherever the customer puts it. For SaaS, Nuix's Security Requirements commit to logically separated databases and file shares dedicated to each customer, with encryption at rest and in transit. No hosting provider, region or processing location for the SaaS offering is stated.

The security page shows marks for ISO/IEC 27001:2022, HIPAA, GDPR compliance and an AICPA SOC logo. Each is an image with no text claiming certification, naming a certifying body or giving a period or scope, and one image file name misspells the statute it refers to. Nothing in the page text says any certification is held. Annex II of the published agreement lists regular benchmarking and testing against industry standards, giving ISO 27001, the SANS Top 20 controls and NIST guidelines as examples, on certain systems, which is benchmarking rather than certification. A trust page exists at a published subdomain, named twice in the agreement for backup and disaster recovery detail and subprocessor change notices.

The Nuix Discover page states that the product holds ISO 27001 and SOC 2 certifications. Nuix's Security Requirements of June 2026 bind it, for SaaS offerings, to obtain an annual third party audit report on ISO/IEC 27001 compliance covering the cloud infrastructure, client and web applications, APIs and software development. They also require AES 256 encryption at rest and in transit and logical separation of each customer's databases and file shares. No auditor, coverage period or route to a report is published, and Nuix runs no public trust center.

The platform page speaks of accurate AI algorithms and AI-powered analytics, and the addendum's own definition of the services includes the development of future functions, features, tools, algorithms and models, so a model layer exists. No model is named, no version, no provider and no hosting arrangement for the model layer, and no commitment to notify customers when any of it changes is published. Clause 2.4 states that the services may use technology and data licensed by Augmetec from third parties, sublicensed to the customer, and names none of them. One infrastructure provider is named and is not a model: Google Cloud Platform, identified in Annex II as the hosting environment and as the source of the secret management service. The agreement refers to an approved subprocessor list at Annex III of the published PDF.

The supply chain is partly disclosed. The vendor describes proprietary AI in the Engine and custom language models trained for the customer's use case. A bring your own AI option lets the customer run its chosen models inside its own environment, which places the third party model, where it runs and who controls it with the customer. A named customer describes using the latest AI advances within its own environment on that basis. Nuix does not publish the name of any model it supplies, any provider behind its own models, or a commitment to notify customers of changes.

There is no pricing page in the site navigation and every route is a demo request. The agreement supplies the shape: the software is licensed under an Order Form as a Package with stated features and pricing, User Rights itemize the number of user licenses and the software instance subscribed to, each authorized user holds a single account that cannot be shared, and the commitment runs across an Initial Term with defined Renewal Terms. That makes it per user licensing on a committed term, with the package defined in an order form. No rate, band, minimum or implementation cost is published. The one published number is a comparative claim rather than a price: the platform page states that the product is up to eighty percent cheaper than other platforms, naming no comparator and no basis.

A unit and structure are published without a figure. The EULA effective 30 September 2025 remains published for agreements made under it. It sells a consumption license at one terabyte of uncompressed data processed per year, aggregated across instances and expiring unused at the end of each term. Usage is reported quarterly and reconciled at the overage rates in the order form, renewals run for twelve months with ninety days' notice, and premium support costs extra. The Solutions Terms of July 2026 leave the usage entitlement to the order. No figure, plan name or price page is published.

The product covers internal and regulatory investigations end to end, from scoping through evidence management, witness interviews and analysis to reporting, extending to ethics, compliance and regulatory reviews. The addendum's data description names identity documents, invoices, expense receipts, financial payment data and material from corporate and ephemeral messaging apps, and the founder's background covers fraud, money laundering, bribery and corruption. Users are law firms and in house professionals. No size band is stated, no jurisdiction is named despite an English law agreement and a UK base, nothing says which investigation types or regulated sectors the product does not suit, and nothing covers languages for a platform taking in messaging data in many languages.

The Neo Legal page lists law firms managing ediscovery at large scale, corporate legal teams handling regulatory matters and disputes, financial services firms, compliance officers, investigation teams and security leaders. Separate solution lines cover investigations and data privacy, with industry pages for telecommunications, integrity agencies and financial regulators. The Engine handles more than a thousand file types. The territory clause and the choice of installation on premises or SaaS set the boundaries, and no matter type or practice area is named as unsupported.