X
X1 Search

X1 Search is a Windows desktop application from X1 Discovery, the Pasadena company behind X1 Enterprise Platform. It finds a person's own email, attachments, files, chats and cloud content from one window as they type. Sources include Outlook and Exchange mail with calendars and contacts, archived Microsoft 365 mailboxes, Teams chats and channels, SharePoint and OneDrive. PST files, file shares, Slack, Gmail, Google Drive, Box and Dropbox are covered too, with more than 50 metadata filters, hit highlighting and full previews.

Version 11 adds optional AI that runs on the user's own machine on lightweight Gemma and Llama models. It tags items with categories the user defines and flags personal data such as Social Security, card and passport numbers, and the data never leaves the device. From version 12, a connector lets Claude work over the X1 index and receive only the matching items, never whole mailboxes.

X1 sells to law firms and enterprises, and a 2018 case study describes Sheppard Mullin, an AmLaw 100 firm, rolling it out to more than 1,000 users. A one year license costs $79 and three years $189, with a 14 day free trial and quotes for 50 or more seats.

Capability grades

All 15 axes, graded from public sources on the date shown. Hover a grade to see what the letter means on that axis.

CC on AI CentralityArtificial intelligence is present but peripheral: a feature layer on a product whose value stands without it.

AI Centrality

How much of the product is actually AI. Whether the machine learning is the mechanism the buyer is paying for or a feature layered onto conventional software, and whether the vendor is specific about which is which.

Finding a user's own material fast is the product. X1 builds its own index of a user's mail, files, chats and cloud accounts in place and returns results as the user types, a patented method with no model involved. Version 11 adds optional AI that a user turns on for each data source, tagging items with user defined categories and finding likely personal data with a mix of patterns and models. The version 12 connector lets Claude work over X1's index.

Without the models, indexing, filters, previews, Boolean expressions and actions on results all work as before, which leaves the AI as a feature layer on a desktop finder.

Source: Vendor Published
BB on Citation Accuracy and Hallucination DisclosureGrounding is real and documented, with linked primary sources and a described retrieval method, short of published accuracy figures an outsider can test.

Citation Accuracy and Hallucination Disclosure

Whether the vendor publishes measured accuracy on citations and assertions, grounds output to primary sources, and says plainly what its system does when it does not know. Legal has a documented public record of fabricated citations reaching filed briefs, so an untested claim of accuracy is not evidence.

Results are the user's own emails and files, shown in full preview with matching terms highlighted, so every hit leads back to its source. The AI features label items rather than answer questions. Categorization applies a category set the user defines to each item's own text at processing time. Comprehensive personal data detection runs pattern matching and a local model side by side and deduplicates their findings, and each tag sits on the item it describes.

No accuracy figure or test result accompanies either feature. Through the connector, Claude receives the matching items and writes its own answer, and how that answer cites them is up to Claude.

Source: Vendor Published
BB on Autonomy and Oversight ModelA written commitment that the models work alongside a supervising lawyer, with real review surfaces, short of the full control structure: commonly the threshold at which the system stops or what happens after it is wrong.

Autonomy and Oversight Model

What the system decides on its own, what a lawyer must approve, and whether the vendor documents where the review point sits. A tool that drafts under review and a tool that files without one are different products and different risks.

AI runs only on the data sources a user selects through each source's AI Settings, and administrators can block opting in altogether. Its output is a tag on each item in a column the user can view, sort and filter, never a change to the item itself. Items that lack enough text for a sound decision are tagged UnCategorized and empty ones NoContent rather than forced into a category. A changed category set calls for clearing results and requeuing.

Progress shows as items processed out of the total, and finished items are usable before the source completes. Categorization is meant to spare the user from reviewing every item, and no sample check of the tags is described.

Source: Vendor Published
CC on Operational and Outcome EvidenceCustomer logos and unattributed testimonials stand in for evidence, or results are quoted with no basis stated.

Operational and Outcome Evidence

Named, dated evidence that the product works in production at real firms or legal departments. Case studies with figures and identified customers count. Unattributed testimonials and launch announcements do not.

Sheppard Mullin, then an AmLaw 100 firm with more than 660 attorneys in 16 offices, rolled the product out to more than 1,000 users in a 2018 case study. A knowledge officer at the firm says its lawyers love X1. The company now says nearly half of AmLaw 100 firms deploy or are considering it, and user reviews are collected on SourceForge. Those accounts predate the AI features, and no customer reports a result from categorization, personal data detection or the Claude connector.

Source: Vendor Published
BB on Privilege and Confidentiality PostureSubstantive published commitments on confidentiality and training use, short of the full picture: commonly silence on segregation between users or matters, or on what the underlying model provider may retain.

Privilege and Confidentiality Posture

How client confidences are handled: attorney client privilege and work product treatment, segregation of one client matter from another, whether client data trains any model, and what the vendor commits to in writing rather than in marketing.

Indexes, lookups and results stay on the user's device, and X1 Services send none of them back to X1. The license states that X1 will not obtain the user's data or train its locally run models on it, and the AI privacy policy adds that content is not used to train, fine tune or improve any model. Optional outside AI services from providers such as Anthropic, OpenAI, Google or Microsoft receive only the items a user sends and fall under those providers' terms.

The connector is described as permission aware, keeping each user to what existing access controls allow. Privilege and work product are not addressed by name.

Source: Vendor Published
CC on UPL and Professional Responsibility PostureA boilerplate disclaimer sits in the terms while the marketing describes the product in advice terms, or the intended audience is left ambiguous.

UPL and Professional Responsibility Posture

Whether the vendor is clear that it supplies a tool rather than legal advice, who its audience is, and how it addresses unauthorized practice of law, competence and supervision duties, and jurisdiction limits. ABA Formal Opinion 512 is the reference point. Where the advice line is not the duty a product raises, the axis is read through the nearest professional duty it does raise: judicial conduct rules and the reviewing duty for products sold only to courts, and the duty to bill for time actually spent for products that draft time entries.

Nothing in the product gives legal answers: it finds and organizes material, and its license delivers the software as is. The marketing covers business and personal use alongside law firms, and nothing speaks to a lawyer's duties when firm or client data passes through the Claude connector to an outside assistant. The license also states that the software is not designed to preserve evidence or retain data, so it is not sold for legal holds or collection, which X1 offers through its Enterprise Platform.

Source: Vendor Published
CC on AI Governance and Bias DisclosureResponsible AI principles are published without a mechanism, a testing regime, or anything a buyer could audit.

AI Governance and Bias Disclosure

Published governance over model behavior: who owns it inside the vendor, what is tested before release, and what is disclosed about disparate output across matter types, parties, or populations.

Model use is left in the customer's hands. X1 ships frozen Gemma and Llama models, runs them only on the endpoint and collects no prompts or chat histories, and administrators can stop users from opting in to AI features through Windows Group Policy. An AI benchmark runs a quick local test and rates the AI experience a given computer can expect. No owner for the models, release testing regime or check of how categorization and personal data detection perform across content types, languages or document lengths is described. Text past a configurable length is cut before the comprehensive personal data scan runs.

Source: Vendor Published
BB on AI Safety and Data StewardshipSubstantive published policy covering most of the ground, short of the full set: commonly no named subprocessor list or no stated incident practice.

AI Safety and Data Stewardship

Retention, deletion, access control, and what happens to prompts and documents after they are processed. Whether the vendor states its subprocessors and its incident practice, or leaves the buyer to assume.

Customer content is not collected, accessed or stored by X1, apart from material a user shares for support. Indexes, a compressed local cache of extracted text and AI tags all sit on the user's own disk, where the user can clear categories or turn the cache off. Requests and results passed through the API or the Claude connector are handled in transit and not kept after the request. Diagnostic data and error logs are kept for up to 24 months, and optional error reports can include indexing status and the user's recent lookups.

Data sent to an outside AI service follows that provider's retention. No named subprocessor list or incident notice practice is offered.

Source: Vendor Published
CC on AI Liability and RecourseLiability is addressed only through a standard limitation clause that disclaims the exposure the product creates.

AI Liability and Recourse

What the vendor stands behind contractually when its output is wrong. Indemnities, caps, carve outs, insurance, and whether any of it is published or only reachable through a negotiated agreement.

The license delivers the software as is, with no warranty of any kind, and caps X1's liability at the greater of 10% of the amount paid for the product or $100. The user indemnifies X1 for use outside the agreement, no intellectual property indemnity runs from X1 to the user, and the user waives claims against X1 over third party AI services connected to the application. California law governs the license, revised March 20, 2026. Its terms also say the software is not designed to preserve evidence or retain data.

Source: Vendor Published
BB on Practice Systems Integration DepthReal integrations exist and are documented, short of depth: named connections without a description of what they actually move.

Practice Systems Integration Depth

How deeply the product reaches into the systems legal work already lives in: document management such as iManage and NetDocuments, Word and Outlook, contract lifecycle management, matter management, e-billing, and court filing systems.

Indexes cover Outlook and Exchange mail with attachments, calendars and contacts, archived Microsoft 365 mailboxes, Teams chats and channels with their attachments and reactions, and SharePoint and OneDrive. They also reach PST files, file shares, Slack, Gmail through an API connector, Google Drive, Box and Dropbox, and users act on results inside the interface. Version 12 adds a plugin that gives Claude a local MCP server over X1's index, installed from the application's MCP folder in about a minute.

Claude Team and Enterprise administrators can install it by default for their organization, and it works in tasks run from the Claude desktop app and in Claude Code, not in ordinary chat. No connector reaches iManage, NetDocuments, a practice management system or a billing platform.

Source: Vendor Published
BB on Deployment Model and Data ResidencyDeployment model is stated clearly with partial residency detail, or residency is offered without the processing location being addressed, or the tenancy model is stated on its own with no residency detail published.

Deployment Model and Data Residency

Where the software runs and where the data sits. Multi tenant cloud, single tenant, private deployment, on premises, and whether region of residence is a published option or an enterprise conversation.

The application installs on Windows 11 and indexes in place on the user's machine, with AI computation also on the device. It runs on the CPU alone, more slowly, and uses Nvidia graphics through CUDA or AMD graphics through Vulkan where present. X1 states that data does not leave the endpoint and is not captured by X1 for any purpose. The Claude connector needs Windows 11 with .NET Framework 4.8 or later, and Windows 10, 32 bit systems, Snapdragon ARM processors and Macs are not supported. Enterprise licensing is arranged with sales, and what changes in a managed or virtual desktop rollout is not described.

Source: Vendor Published
DD on Security Certifications and Trust CenterNo independent security attestation located.

Security Certifications and Trust Center

Independent attestation a buyer can pull without a sales call: SOC 2, ISO 27001, penetration test summaries, a trust center with current reports and named scope rather than a badge image.

No security certification, audit report or trust page covers the product, and the privacy policy refers to standard security procedures without naming a standard. Installation phones home to authenticate the license. Protection rests on the design rather than an attestation: indexes and AI processing stay on the endpoint, and administrators can block AI features by Group Policy.

Source: Operator Verified
BB on Model Supply Chain DisclosureThe supply chain is partly disclosed: providers named without change notification, or architecture described without the providers.

Model Supply Chain Disclosure

Which models sit underneath, whose they are, where they run, and whether the vendor commits to telling customers when that changes. A legal buyer inherits every dependency it cannot see.

Frozen language models from Google's Gemma and Meta's Llama families run locally for categorization and personal data detection, included with version 11. Optional outside AI services from providers such as Anthropic, OpenAI, Google or Microsoft receive only what a user directs to them. The connector hands matching items to Claude through a local MCP server and an /x1 skill that teaches Claude to work with file paths rather than pull in large attachments. Model versions and sizes are not given, and there is no notice when X1 changes the bundled models.

Source: Vendor Published
BB on Commercial TransparencyReal pricing is published for part of the range, with enterprise tiers withheld, or the unit and structure are stated without the figure.

Commercial Transparency

Whether a buyer can learn what this costs without entering a sales process: published rates, the unit being charged, what sits behind an enterprise tier, and what implementation adds.

A one year license costs $79 and a three year license $189, against a list price of $237, each with support and updates for the term. Every version 12 license includes the Claude connector at no extra charge. Licenses are bought online by card in USD through FastSpring, and a 14 day free trial needs no sales call. Purchases of 50 or more licenses, like enterprise licensing, are quoted by sales. Renewal runs by automatic payment or another arrangement with X1, and current license holders can upgrade to version 12.

Source: Vendor Published
CC on Firm and Practice CoverageCoverage is claimed broadly, for all firms or all practice areas, without evidence that the breadth is real.

Firm and Practice Coverage

Who the product is actually built for. AmLaw, midlaw, small firm and solo, in house departments, government and courts, and which practice areas are supported rather than merely claimed.

Law firms and corporate legal teams sit beside business and personal users in X1's marketing, and the case study names a large firm's attorneys and associates as the users. The product finds any kind of content rather than serving a practice area, and its examples run from archived and spam email to Slack and Teams conversations. It runs only on Windows 11 PCs, and no limit by firm size or matter type is stated. Volume buyers of 50 seats or more are routed to sales, while the wider X1 platform serves legal, compliance, governance and investigation teams at enterprise scale.

Source: Vendor Published
Sources on file

8 public documents

The public pages on file for X1 Search, with the recorded signals each one supports and the date it was last read. Open any of them and check the reading against the record.

Pricing

$79/year per license

USD, as published, never converted
  • A one year license costs $79.
  • A three year license costs $189, against a list price of $237.
  • Both include support and updates, and a 14 day free trial is available.
  • Orders of 50 or more licenses are quoted by sales.

The application is sold as a Windows term license, for one year at $79 or three years at $189, each with support and updates for the term. Licenses are paid online by card in USD through FastSpring, and every license from version 12 includes the Claude connector at no extra charge. A 14 day free trial is available. Purchases of 50 or more licenses and enterprise licensing are quoted by sales.

Implementation: None. The user downloads and installs the application.

Note: A term license ends at expiry unless renewed by automatic payment or another arrangement with X1. The license sets no refund terms.

Legal Signals

What each signal means

A signal records what public sources say on the date shown. It is not a grade and it is not a recommendation. Where a signal reads Not addressed, it means the index did not locate the material in public sources on that date, which is a statement about disclosure rather than about the product.

Confidentiality and Privilege

Client Data in Training

Can material a lawyer puts into this product be used to train a model?

Never, in the contract

The published terms prohibit training on customer content. Not a policy page, the agreement.

The end user license, revised March 20, 2026, states that X1 will not obtain the user's data or further train the locally run models with it, and the AI privacy policy adds that content is not used to train, fine tune or improve any model. Outside AI services a user connects follow their own terms.

Source: Vendor Publishedwill not obtain any of YOUR data or further train these LLMsAs of Nov 20, 2026Evidence

Prompt and Output Retention

How long does the product keep what a lawyer typed, and can that be set to zero?

Customer controlled, no zero option

The customer controls the retention window, by product configuration or by contractual instruction, but zero retention is not stated as available.

Indexes, content caches and tags sit on the user's own machine for as long as the user keeps them. The local cache of extracted text stays until the user clears it or turns it off. Requests and results passed through the API or the Claude connector are not kept by X1 once a request completes, and diagnostic logs stay for up to 24 months.

Source: Vendor Publishednot retained by X1 after the request has been completedAs of Nov 20, 2026Evidence

Ethical Walls and Matter Segregation

Does retrieval respect the firm’s ethical walls, or can the model read across them?

Not addressed

No located public material addresses walls or matter level segregation.

Each installation indexes the sources its user signs in to, and Claude sees only what that index holds. The connector is described as permission aware, respecting existing access controls. Nothing describes walls between matters or clients within a user's data or across a firm's seats.

Source: Operator VerifiedAs of Nov 20, 2026Evidence

Third Party Request and Subpoena Notice

If someone subpoenas the vendor for a firm’s data, does the firm hear about it first?

Disclosure addressed, notice absent

Published terms or policy address disclosure to authorities or in response to legal process, and no commitment or reservation regarding customer notice is located anywhere. The vendor has told the customer that data can leave and has said nothing about whether the customer hears of it.

Personal information may be disclosed if permitted or required by law, including in response to a warrant, subpoena or other legal process, with no promise to tell the user. Indexed content stays on the user's machine and does not reach X1.

Source: Vendor Publishedif permitted or required by lawAs of Nov 20, 2026Evidence
Accuracy and Authority

Primary Law Corpus Provenance

Where does the law in this product come from, and does the vendor have the right to use it?

Not addressed

No located public material identifies the corpus behind the product’s answers.

X1's index holds the user's own email, files, chats and cloud accounts, with no collection of statutes or case law behind it. Its coverage is set by the sources the user connects and the scans the user schedules.

Source: Operator VerifiedAs of Nov 20, 2026Evidence

Good Law Verification

Does the product tell you when the authority it just cited has been overruled?

Not addressed

No located public material addresses whether authority is checked for subsequent history.

No legal authority is retrieved, and nothing tests whether a case or rule still stands, since the product works over a user's own mail, files and chats.

Source: Operator VerifiedAs of Nov 20, 2026Evidence

Refusal and Uncertainty Behavior

What does the product do when the answer is not in the corpus?

Documented

The vendor describes refusal or abstention behavior in public materials.

Categorization tags an item UnCategorized when it lacks enough text or context for a reasonable decision, and empty items are tagged NoContent, instead of forcing either into a category. Items still waiting are tagged ToBeProcessed, and the comprehensive personal data scan cuts text past a configurable length before the model reads it.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Fabricated Citation Record

Does a public court record exist addressing fabricated or hallucinated legal citations in output from this product?

None located

No court order, opinion or disciplinary record addressing fabricated or hallucinated legal citations produced by this product has been located as of the date shown. This is a statement about the public record on that one subject, not a finding about the product, and this signal is not a litigation history.

The AI Hallucination Cases database kept by Damien Charlotin has no entry naming X1 or X1 Discovery. The product finds and labels a user's own documents rather than drafting filings.

Source: Operator VerifiedAs of Nov 20, 2026Evidence
Professional Responsibility

Bar Guidance Alignment

Has the vendor engaged in public with the ethics opinions its buyers are bound by?

Not addressed

No located public material engages with bar or ethics guidance.

Ethics rules and bar guidance on lawyers' use of AI go unmentioned, including for the Claude connector that sends matching firm documents to an outside assistant. The controls a firm gets are per source opt in for the local models and an administrator block through Group Policy.

Source: Operator VerifiedAs of Nov 20, 2026Evidence

Billing and Fee Posture

Does the vendor address what happens to the bill when the work takes an hour instead of six?

Savings claims only

Public materials claim time savings without addressing billing or disclosure, and the product sits inside a fee relationship between a lawyer and a client where those savings would change the bill.

X1 promotes faster retrieval of email and files for lawyers at firms such as Sheppard Mullin, Gmail lookups up to three times faster through its API connector, and lower Claude token costs from sending only matching items. Nothing covers how the time saved or AI assisted retrieval appears on a client bill.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Outside Counsel Guideline Readiness

Can a firm get this vendor through a client’s AI clause without a bespoke negotiation?

Subprocessors listed

A current subprocessor or model provider list is published.

The AI privacy policy, updated August 18, 2026, names the model families shipped with the product, Google's Gemma and Meta's Llama, and lists the services the software connects to (Microsoft, Google, Dropbox, Box, Yahoo and AOL) with links to their terms. No client facing disclosure pack is offered.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Court Disclosure Support

If a judge’s standing order requires an AI disclosure, can the product produce one?

Not addressed

No located public material addresses court disclosure or verification certification.

No record of AI use for a court disclosure is produced, and the license states the software is not designed to preserve evidence or retain data. Tags and personal data flags live in the user's local index rather than in an exportable log.

Source: Operator VerifiedAs of Nov 20, 2026Evidence
Contact

Correct a record, or ask how something was graded

Every grade and every signal on this index is drawn from public sources and dated. If a record is wrong, out of date, or missing an artifact the index did not locate, send the source and it will be reviewed and the record redated. Vendors are welcome to submit documentation. Nothing on this index is for sale, including a listing, a placement, or a grade.

AI Legal Index

The AI Legal Index is an independent index that tracks changes to AI vendors in legal. It holds 617 vendors across 9 categories, each graded on the same 15 capability axes and recorded against 12 legal signals, from public sources. No vendor pays for inclusion, placement, or rating.

Index Status
Last index update
November 20, 2026
The AI Legal Index is an editorial reference. It is not a regulatory body, not a law firm, and nothing published here is legal advice or a recommendation to retain or avoid a vendor. Records are verified against published sources, bar guidance and public court records. Where a record reads not addressed, the material was not located in public sources on the date shown. See the Methodology page for evaluation standards and limitations.
© 2026 AI Legal Index
3801 N Capital of Texas Hwy, Ste E240 · Austin, TX 78746