A
Aid4Mail Investigator

Aid4Mail Investigator is the email forensics and eDiscovery edition of Aid4Mail, Windows software from Fookes Software Ltd of Charmey, Switzerland, which has made email tools for more than 25 years. It collects mail from Gmail, Microsoft 365, Yahoo and other accounts and from more than 40 file formats, including PST, OST, MBOX and Apple EMLX. Filters by keyword, participant and date run at the server before download, which the company says cuts data by 90 percent at source.

It recovers deleted and corrupted email, carves mail from unallocated space, removes duplicates and collects cloud attachments from OneDrive, SharePoint and Google Drive. Boolean, proximity, wildcard and regular expression matching narrow the set, and output goes to PST, PDF, HTML with a portable tagging viewer, or formats for FTK, EnCase, Nuix, Intella Pro and RelativityOne. Version 6 added AI Filter, Classify and Analyze, which apply plain language instructions to judge responsiveness, privilege and hot documents and to summarize, translate and identify entities.

They run on the customer's own key for Anthropic, Google, OpenAI and other cloud models or on enterprise platforms such as Amazon Bedrock and Microsoft Foundry. Offline models run through Ollama or LM Studio, and ambiguous emails are labeled INCONCLUSIVE for human review. Investigator costs $129 a month or $999 a year for one user, and Fortune 500 companies, law enforcement agencies and law firms including Seyfarth Shaw are among its users.

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.

BB on AI CentralityThe models are the engine of a core capability, layered on a product that would still function without them as a document or workflow system.

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.

AI Filter, Classify and Analyze drive responsiveness, privilege and hot document review, the core eDiscovery task in this edition. Classify sorts mail into Responsive, Privileged, Hot or custom categories, and Analyze summarizes, translates, identifies entities and infers signals such as sentiment or deception.

Beneath them sit collection, conversion, recovery, deduplication and filtering across more than 40 formats that the product performed for years before AI was added, and the Converter edition still sells without it.

Source: Vendor Published
AA on Citation Accuracy and Hallucination DisclosureMeasured accuracy is published with the test set described and the failure modes named. Output grounds to primary authority the reader can open, citation status is checked, and the system states when it found no support.

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.

Test 1 ran 15 cloud and offline models over 2,000 emails at 6 percent responsive prevalence, with recall from 99.17 to 100 percent and F1 from 90.23 to 99.59 percent. Twelve models missed no responsive email, and three missed one of 120.

Test 2 measured exact match accuracy across eight languages and counted INCONCLUSIVE replies and errors as failures. Results are said to depend on prevalence, prompt quality, corpus and model. Ambiguous emails are labeled INCONCLUSIVE rather than forced into a category. Both tests are the company's own, and the kits let a buyer rerun them.

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.

Users test a prompt with the Verify function and run a representative sample before full processing. Emails the model cannot place are labeled INCONCLUSIVE and sent to human review, and reviewers can pull up every Responsive or INCONCLUSIVE email in the portable viewer and tag as they go.

The design keeps a human in control of every consequential decision, and for sanction bearing decisions AI output is meant to support the reviewer's judgment rather than replace it. Lawyer sign off on a production set and the response to a wrong call are not addressed.

Source: Vendor Published
BB on Operational and Outcome EvidenceReal deployment evidence with substance, short of full attribution or measurement: a named customer without figures, or figures without the named customer.

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.

Richard Lutkus, a partner at Seyfarth Shaw, counts it among his must have tools, and Jon Hanna of Advanced Discovery finds it at least twice as fast as his other tools. Gregory Kerr of Lubin Olson & Niewiadomski and the eDiscovery commentator Craig Ball also vouch for it.

Fortune 500 companies and law enforcement agencies in more than 100 countries are named as users. Reported results, such as 90 percent data reduction at source and an internal pilot at 1.78 emails a second, carry no named customer, and the named praise carries no method or date.

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.

The EULA states that the software uses data only to deliver its functions and does not store or share it with Fookes Software, and offline models keep email content on the customer's hardware. With cloud models, email text and every attachment filename and extension go to the chosen provider under its terms, and extracted attachment content goes too when Include Attachment Data is on.

Likely privileged emails are flagged for attorney confirmation, and provider retention and training are left to the customer to verify.

Source: Vendor Published
BB on UPL and Professional Responsibility PostureA real position is published on advice versus tooling, short of full treatment: commonly a disclaimer without the supervision and competence dimension, or silence on jurisdiction limits.

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.

AI output is positioned as support for the reviewer's judgment on sanction bearing decisions, privilege calls go to attorney confirmation, and the legal section of the AI guide is labeled general guidance rather than legal advice. Users are told to consult legal professionals on compliance and to check their approach with privacy and forensics experts.

Proportionality and the lawyer's supervision and competence duties are not addressed.

Source: Vendor Published
BB on AI Governance and Bias DisclosureA published governance framework with real substance, short of testing results or a named owner.

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.

Two benchmarks measure the classifier across models. Both ship as downloadable kits with the prompt file, session template and full email corpus, so a buyer can rerun them on its own provider and hardware. Test 2 ran 1,600 synthetic emails in eight languages, from English and French to Korean and Japanese, each judged by exact match to its label.

Accuracy ranged from 81.50 to 99.38 percent among retained models, and the four weakest lost the most ground in Korean. Even at temperature 0 the models are not fully deterministic, so reruns can differ. No accountable owner, release testing regime or governance framework stands behind the benchmarks.

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.

Processing runs on the customer's machine, temporary files are deleted when a job completes, and OAuth tokens are encrypted with AES 256. Fookes holds only activation data, on Swiss servers.

Users are told to process the minimum personal data, delete or anonymize data once it is no longer needed and put a DPA in place with any cloud AI provider. Incident practice is not stated, and what a provider keeps depends on the customer's own agreement with it.

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 EULA for Aid4Mail 6, updated June 5, 2025, provides the software as is, limits liability to the amount paid for it and contains no indemnity. Nothing stands behind a misclassified email.

AI tools are described as able to make mistakes, and Analyze output is meant to be reviewed before anyone relies on it.

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.

Output feeds FTK, EnCase, X-Ways, Belkasoft X, AXIOM, Intella Pro, Nuix and RelativityOne, and collection reaches Gmail, Microsoft 365, Yahoo, OneDrive, SharePoint and Google Drive. Python filter and modifier scripts act on AI results for routing, redaction and reporting, and a public GPT from Fookes writes prompts or scripts from a plain language description.

Load file specifications for each review platform, and any return path from them, are not described.

Source: Vendor Published
AA on Deployment Model and Data ResidencyDeployment options and data residency are published, including the regions available, what changes between tiers, and where processing happens as distinct from where data is stored.

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 software installs on Windows desktops, laptops, servers or a shareable USB drive, and email stays on the customer's hardware. Offline models through Ollama or LM Studio keep AI processing local. Cloud processing goes to the provider and region the customer chooses, including an EU Gemini endpoint and enterprise platforms with regional residency.

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 independent certification or audit covers the software or Fookes Software. Stored OAuth tokens and sensitive data are encrypted with AES 256 and traffic with TLS 1.2 or higher.

Email is processed on the customer's own machines rather than on a Fookes service, and the release in use is Aid4Mail 6.3.2, dated July 2, 2026.

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.

Cloud providers are Anthropic, Google, Meta AI, Mistral AI, OpenAI and xAI, and the enterprise platforms are Amazon Bedrock, Gemini Enterprise Agent Platform and Microsoft Foundry. Offline models include Ministral 3 14B, Mistral Small 3.2 24B, Magistral 24B, Qwen 3.8 27B, Gemma 4 and Llama 3.3 70B, run through Ollama or LM Studio.

The customer picks the model and holds the key. Cloud providers can revise a model without notice, so a pinned cloud version or an offline model is recommended for long running matters, and Fookes makes no change notice commitment of its own.

Source: Vendor Published
AA on Commercial TransparencyA buyer can learn what this costs without entering a sales process: published rates, the unit being charged, and what implementation adds.

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.

Investigator costs $129 a month or $999 a year for one user, Converter $49 or $299, and Enterprise $649 a month with a four month minimum or $4,999 a year for unlimited internal users. AI use is paid directly to the chosen provider, and the Test 1 run cost $0.67 to $11.03 per 2,000 emails across the eight cloud models.

Prices exclude VAT and GST, annual renewals get 10 percent off, updates are free during a subscription and the trial has no time limit.

Source: Vendor Published
BB on Firm and Practice CoverageSegment and practice coverage is described with substance, short of the boundaries: what is supported is clear, what is not is left open.

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.

Buyers are eDiscovery providers, law firms, corporate investigators, forensic examiners and law enforcement, for email and attachments only. More than 40 formats and the major webmail services are covered.

Language reach depends on the model, with Gemini covering more than 35 languages, and Korean was the weakest of eight tested languages for the four weakest models. Mail systems handled poorly are not stated.

Source: Vendor Published
Sources on file

2 public documents

The public pages on file for Aid4Mail Investigator, 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

From $84/user/month billed yearly ($999/year)

USD, as published, never converted
  • Aid4Mail Investigator costs $129 a month or $999 a year for one user.
  • AI analysis is included, and AI usage is paid directly to the provider the customer picks.
  • Offline AI models cost nothing extra to run.
  • Annual plans get 10 percent off at renewal, and the free trial has no time limit.

Investigator is licensed per user as a subscription, at $129 a month, cancel anytime, or $999 a year, which works out to about $84 a month. AI Filter, Classify and Analyze are included, with AI usage paid directly to the cloud provider the customer chooses and no marginal cost for offline models. Processing of accounts and files is unlimited, and Investigator runs accounts concurrently. Converter, without forensic or AI features, costs $49 a month or $299 a year, and Enterprise costs $649 a month with a four month minimum or $4,999 a year for unlimited internal users. Updates are free during the subscription, and virtual machine licensing is priced on request.

Implementation: None stated.

Note: Prices exclude VAT and GST. Annual subscriptions get 10 percent off at renewal. A free trial needs no card and has no time limit.

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?

Purpose limited, in the contract

The customer agreement or data processing addendum contractually limits use of Customer Data to providing the contracted service, and no surface names training either way. The limit is bound, which a policy page is not, but it is not an express training prohibition. If any surface names training in either direction, one of the other values is true and this one is not.

The EULA limits the software's use of data to delivering its functions and states that nothing is stored or shared with Fookes Software. Training by a cloud AI provider depends on the provider the customer chooses. Users are told to put a DPA in place and review its terms.

Source: Vendor Publishedaccesses your data solely for the purposes of providing the intended functionalityAs of Nov 18, 2026Evidence

Prompt and Output Retention

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

Customer set, zero available

The customer sets the retention window and no retention is an available setting.

Email and AI output stay on the customer's hardware, temporary files are deleted when a job completes and Fookes keeps only activation data. Offline models retain nothing outside the customer's environment, and cloud retention follows the chosen provider's terms.

Source: Vendor Publisheddoes not store or share your data with Fookes SoftwareAs of Nov 18, 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.

Cases run as separate local jobs on the customer's machines, and no user or matter permission model is described.

Source: Operator VerifiedAs of Nov 18, 2026

Third Party Request and Subpoena Notice

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

Not addressed

No located term or policy addresses third party requests for customer data.

Fookes holds no customer email, and its privacy policy, updated October 29, 2026, does not address requests from authorities.

Source: Operator VerifiedAs of Nov 18, 2026
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.

The AI classifies the customer's own email, and no external legal corpus is involved.

Source: Operator VerifiedAs of Nov 18, 2026

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.

The product reviews email and does not cite legal authority.

Source: Operator VerifiedAs of Nov 18, 2026

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.

Emails the model cannot place are labeled INCONCLUSIVE and sent to human review, and the label is added automatically when a user supplies a category list without it. Test 2 counts INCONCLUSIVE replies and errors as failures in its accuracy scores.

Source: Vendor PublishedAs of Nov 18, 2026

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.

No court decision in the AI Hallucination Cases database maintained by Damien Charlotin names Aid4Mail.

Source: Operator VerifiedAs of Nov 18, 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.

No professional conduct opinion on AI use, such as ABA Formal Opinion 512, is cited. The legal section of the AI guide is labeled general guidance rather than legal advice and sends users to legal professionals.

Source: Operator VerifiedAs of Nov 18, 2026

Billing and Fee Posture

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

Audit record only

A usable record of AI assisted work exists with no published fee guidance.

Processing logs record which emails AI Filter passed or rejected, and labels and tags export with the output. AI cost per run is measured, at $0.67 to $11.03 per 2,000 emails in Test 1, but how AI review is billed to a client is not addressed.

Source: Vendor PublishedAs of Nov 18, 2026

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.

Every supported AI provider, platform and offline model is named, and the privacy policy lists the hosts and services Fookes uses for its website, licensing and support. No customer email passes through Fookes.

Source: Vendor PublishedAs of Nov 18, 2026

Court Disclosure Support

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

Partial record

Some elements of the record are available, short of a document level export.

Audit logs cover all AI processing, hashes are preserved, and outputs export as PDF, HTML, CSV, TSV, XML or JSON. The export made at the time of a run is treated as the controlling record, and users archive the prompt, model, provider, region, settings, run time and logs by hand because reruns can differ.

Source: Vendor PublishedAs of Nov 18, 2026
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 535 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 19, 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