Intaker vs Perspective AI: how they compare in 2026
Intaker and Perspective AI both sell intake conversion, and they take opposite routes to it. Intaker runs a scripted chat on more than 1,400 authored prompts, with a lawyer's recorded video playing inside the conversation. Perspective AI generates the conversation, adapting on the answers and structuring what it hears into the fields a form would have collected. What separates them is what each is for. In Perspective AI the model is the entire product, and removing it leaves nothing. Removing Intaker's models leaves a working chat widget, a CRM and a follow up engine, which is most of what a firm buys. Intaker's advantage is where the lead lands. It names Clio and Filevine alongside Salesforce, Calendly and Google My Business, while Perspective AI names only Slack and describes its other destinations as an inbox or a CRM.
At a glance
All 15 axes, side by side
The same grid applied to every vendor in the index, graded from public sources. Hover a grade to see what the letter means on that axis.
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.
Intaker is a scripted chat and CRM product with AI capability layered across it. The chat runs on more than 1,400 custom prompts and prebuilt scripts for different practice areas, which is written conversational content rather than generated conversation. The surrounding platform, lead management, texting, follow up sequences, virtual inbox workflows and reporting, is fixed automation. The stated AI capabilities are real and specific: natural language processing, language detection and automated responses, and both the vendor and independent directories describe the intake automation as AI driven. Removing the models would leave a working scripted chat widget, CRM and follow up engine, which is most of what a firm buys.
The model is the entire product. Adaptive conversation, follow up on vague answers and extraction that turns unstructured answers into structured fields are what the vendor sells. Remove the model and nothing remains, because the thing being replaced is the form. Perspective AI is a horizontal intake platform serving healthcare, financial services, insurance, education, real estate and internal operations as well as legal. Its legal offering is specific: a dedicated legal use case names personal injury, family law, estate planning and immigration qualification, the vendor publishes an analysis of legal intake software that places itself in that market, and it describes capturing the parties and matter details needed to flag conflicts. Legal is one of several verticals rather than the company's center of gravity.
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.
Nothing is published. A chat running on 1,400 written prompts produces less free text than an adaptive conversational agent, so the usual hallucination exposure is smaller; that is a design property, not a disclosure. Nothing addresses what the AI layer does: no accuracy figure for language detection, response selection or qualification, no error rate and no evaluation. Nothing says what happens when a prospective client's message matches no written prompt, which is where a scripted system either falls back gracefully or answers wrongly.
Nothing published, and the extraction claim is the thing that needed measuring. The vendor's central technical proposition is schema aligned data extraction, structuring unstructured conversational answers into defined fields, and it markets the output as validated structured outputs on which downstream automation triggers rather than guesswork from dropdowns. Extraction accuracy is directly measurable and no figure is published: no field level accuracy, no error rate, no evaluation, no statement of what happens when a conversational answer is ambiguous and the schema requires a value. The word validated is doing significant work in the marketing and nothing states what validation means or who performs it. The consequence is specific: a misextracted field in legal intake produces a structured summary that reads as authoritative and routes a matter incorrectly, and the conversation it came from is the only place the error is visible.
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.
Two published mechanisms put a human into the conversation itself. Live Call Connect, a click to call widget, turns a chat lead into a live phone transfer to the firm, so the prospective client reaches a person while still engaged rather than after a record is handed off. The video feature works differently: an attorney records intro and outro videos that play in the chat, so a named person is visibly present in an automated conversation. Nothing is bounded. There is no statement of what triggers a call transfer or whether only the caller can start one, and no threshold on qualification. Nothing says what the chat does outside business hours when no one can take a transfer, and there is no override or review path for a qualification decision.
Two published mechanisms put a person at the right points. First, configuration can be reviewed before deployment: the firm sets routing rules for strong and weak fits and can preview the full conversation before it goes live, so a lawyer can inspect exactly what the system will say to a prospective client rather than finding out afterward. Second, on conflicts, the vendor states that the platform captures the parties and matter details during the conversation and routes the summary to staff who confirm against the firm's own records. That is an explicit statement that the machine gathers and a person decides on the one question where an automated determination would be indefensible. The qualification decision itself is autonomous. Routing strong and weak fits happens without a person, and nothing states what a weak fit outcome does to the inquiry or whether a borderline case is escalated.
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.
Independent listings and verified reviews exist in small numbers, and no customer or outcome figure does. Without relying on the vendor, a reader can find listings on several independent software directories, with provider data stated as verified and reviews moderated for authenticity. There is also a listing in a named legal technology publication, a published starting price and a corporate entity, Intaker, Inc. Directory testimonials are attributed to roles rather than named firms, and volume is thin, with one major platform showing three user reviews. No law firm is named in vendor material, and no case study, usage figure, funding announcement or outcome claim with a figure is published.
Customer evidence is absent in every vertical. No customer is named in any vertical, and no case study, usage figure, funding announcement, independent analyst placement, award or third party review of this vendor was found. The evidence is entirely self authored, though large in volume: an extensive blog publishing category analysis, buyer frameworks and comparative rankings across legal intake, insurance, education, real estate and internal request management, in several of which the vendor ranks itself first. Rankings a vendor publishes about itself are marketing rather than evidence.
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.
Nothing located. No confidentiality statement, no encryption reference, no data handling description and no treatment of privilege or prospective client confidences was found. The product's own design raises the question more sharply than a text chat would: two way business texting means prospective client communications travel by SMS, which is an unencrypted channel outside the firm's control, and the platform centralizes those communications alongside chat transcripts and lead records in a vendor held CRM. Nothing published addresses the confidentiality of any of it, and nothing distinguishes the status of a chat transcript from a prospective client who never becomes a client.
Prospective client confidences have no stated protection. No confidentiality statement, no encryption reference, no data handling description and no treatment of privilege or prospective client confidences was found. The exposure is heightened by the product's own design proposition: the vendor argues that forms lose the context that matters and that its conversations capture the why behind each inquiry, which means the system is built to elicit more sensitive detail from a prospective client than a form would, in named practice areas including family law and immigration. It also operates across healthcare, where the same conversations capture presenting concerns and care needs. A platform designed to extract fuller disclosure publishes nothing about how that disclosure is protected.
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.
No position is published. The chat qualifies prospective clients against criteria for specific matter types using more than 1,400 written prompts, so the system has a substantive exchange about a person's legal problem and reaches a qualification outcome before any lawyer is involved. Nothing says the chat does not give legal advice, describes what automated responses may say when a prospective client asks a substantive question, or tells the person whether they are talking to software or a human. That last point matters here because the video feature presents a named attorney inside an automated conversation. No bar guidance is engaged.
Not located. The system holds an adaptive conversation with a prospective client, asks follow up questions about their legal situation, assesses fit against the firm's criteria and routes on that basis, all before any lawyer is involved. The vendor's own framing is that this happens before the paralegal picks up the phone. That is an automated system holding a substantive discussion about a person's legal problem and reaching a conclusion about it. Nothing published states that the conversation is not legal advice, describes what the agent may say when a prospective client asks a substantive legal question during the conversation, or discloses to the person that they are speaking to an AI. The interaction is conversational and adaptive rather than form based.
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.
Nothing is published about how the models are governed, evaluated or monitored. No AI policy, model card, bias or fairness testing, evaluation method, accuracy monitoring, drift statement, named governance body, ISO 42001 or EU AI Act position is published. The chat qualifies leads automatically, so it decides which inquiries reach a firm, and nothing indicates whether qualification outcomes have been examined across different groups of people. Language detection deserves specific mention: these systems perform unevenly, and a misdetection at first contact affects whether a prospective client who does not speak English can proceed at all. No evaluation of it is published.
Nothing is published on governance. The platform qualifies inquiries against fit criteria the firm defines and routes strong fits differently from weak ones, so it decides which people reach a professional. That includes immigration and family law, where the people being assessed are often in difficulty and have few alternatives. The adaptive design adds to the risk. Because the conversation branches on responses, two people in similar situations can be asked different questions and produce different summaries depending on how articulately they answer early, and nothing indicates whether performance across different groups of respondents has been examined. No AI policy, model card, bias or fairness testing, evaluation, monitoring, governance body, ISO 42001 or EU AI Act positioning is published.
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.
No stewardship position located. Nothing states whether chat transcripts, text message threads, lead records or qualification outcomes are used to train or improve models, no retention period is published, and no deletion right is described. The platform is a CRM as well as a chat widget, so it holds a firm's prospective client base durably by design, and two way texting adds a communications archive on top of it. Nothing published addresses any of that, and no privacy policy or data processing statement is published in the material read.
No stewardship position is published. Nothing states whether intake conversations, extracted data or routing outcomes are used to train or improve models, no retention period is published, and no deletion right is described. The horizontal footprint widens the omission. The same platform captures presenting concerns and care needs in healthcare intake and financial position in wealth management qualification, so one undisclosed training and retention posture covers health information, financial information and legal matter details together. The free tier with no code setup adds to this, since anyone can deploy an agent collecting such material with no published statement about where it goes.
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.
No published position on liability for AI output, warranty, service levels or remedy is published. One exposure is specific to this product: automated follow up sequences send personalized email and text reminders to leads on the firm's behalf. The vendor's system therefore sends outbound messages to consumers, and texting prospective clients engages consumer protection and messaging rules regardless of AI. Nothing published addresses responsibility for the content or sending of those messages, or for an inquiry wrongly screened out, a failure the firm would not see.
No published position is published on liability for AI output, warranty, service levels or remedy. Two exposures are specific to this product. A misextraction produces a structured summary that a firm acts on without seeing the conversation behind it, and the vendor markets automation triggered by those outputs, so an error spreads into routing and workflow before anyone reads it. And a person qualified as a weak fit is routed away without anyone learning it happened. The vendor's own marketing frames the problem it solves as teams moving forward on incomplete information and making weaker decisions, which describes the risk its extraction would create if wrong, and no position addresses it.
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.
Named integrations cover the practice management systems this buyer runs, plus marketing and scheduling. Clio and Filevine are named for practice and case management, Salesforce as the general CRM, Calendly for scheduling, and Google Analytics and Google My Business for traffic and listing attribution. Filevine matters here because plaintiff and high volume consumer firms commonly run it. The Google My Business connection suits an intake product, since many consumer legal inquiries start from a local listing rather than the firm's own site. No integration depth is described for any named system, nothing says whether lead data flows one way or both ways, and no API documentation is published. The integrations are listed, not described as workflows.
Destinations are named by type and no system is named individually. Structured summaries go to an inbox, a CRM or Slack, and the agent can run as a website widget, popup, email link or standalone page, so both the output path and the input surface are described. Slack is the only product named. What a law firm would check is absent. No practice management system is named, there is no Clio, MyCase or Filevine integration or case management connector, and no API documentation is published. A firm cannot confirm that its own CRM is a supported destination or that intake will reach its matter records. A generic CRM destination describes a category rather than an integration.
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.
Nothing located. No hosting provider is named, no region or data residency commitment is published, and no deployment options are described beyond the product being cloud based and embedded on a firm website. The platform holds prospective client chat transcripts, text message history and lead records, and routes text messaging through a carrier layer that is not identified, so neither the location of the data nor the parties handling the messaging can be determined from public material.
Hosting, region and tenancy all go unstated. No hosting provider is named, no region or residency commitment is published, and no deployment or tenancy options are described beyond the places the intake agent can be embedded, which is a placement question rather than an infrastructure one. The gap is wider because the same platform is marketed for healthcare patient intake and financial services qualification, so health and financial data flow through infrastructure whose location is unstated, and both carry obligations that turn on location.
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 certification, attestation, trust center or security page is published. No SOC 2 of either type, ISO 27001, named auditor, penetration testing partner or encryption statement was found on the pages read, and nothing suggests a gated security page exists. The gap matters for a platform holding consumer chat transcripts and text message history for law firms whose own obligations attach to that material.
No certification, attestation, trust center or security page is published. No SOC 2 of either type, ISO 27001, HIPAA statement, named auditor, penetration testing partner or encryption statement was found. The material is absent rather than gated. The HIPAA omission is the notable one. The vendor markets healthcare patient intake that captures history, urgency and care needs, which is protected health information, and publishes no HIPAA posture, no offer of a business associate agreement and no security credential of any kind.
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.
Nothing located. No foundation model provider, model family or version is named, no distinction is drawn between proprietary and third party models, and no subprocessor list was found. The gap is wider than the model layer alone on this product: two way business texting necessarily runs through a messaging carrier or platform, and Live Call Connect necessarily routes through telephony infrastructure, and neither is identified, so a firm cannot determine which third parties carry its prospective clients' messages and calls. The vendor names six integration partners and no processor.
The models behind the product are unidentified. No foundation model provider, model family or version is named, no distinction is drawn between proprietary and third party models, and no subprocessor list was found. The absence is conspicuous against the vendor's own published expertise: it writes detailed comparative analysis of the conversational AI category, classifying competitors by architecture and distinguishing AI native platforms from chat skins on form builders, so it demonstrably understands that the underlying architecture matters to a buyer, and it discloses nothing about its own. A platform conducting adaptive conversations and performing schema aligned extraction is running one or more language models supplied by someone, and no party in that chain can be identified from public material.
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 specific entry price is published by independent sources, and nothing above it. Several independent software directories report a starting price of $80 per month, consistently and with provider data stated as verified by their research teams, and one records that no free trial is available. Nothing describes a tier structure, a unit of charge, what the entry price includes or what raises it. Nothing says whether pricing scales by seats, conversations, leads or messaging volume, which are very different meters for a product that bundles chat, CRM and two way texting. The figure comes from directory listings. No vendor pricing page is published, and one may exist.
A free entry point is published and no price is. The vendor offers a free tier with no code setup and lead capture agent templates, so a buyer can deploy and evaluate the product without a sales conversation. Nothing above it is published. No price, range, tier structure or unit of charge is published, and nothing says whether pricing runs per conversation, per completed intake, per agent, per seat or by volume, which are very different economics as inquiry volume grows. No figures from other sources are published either. The barrier to trying is low and the cost of adopting is undisclosed.
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.
Practice coverage is claimed by volume and described no further. The strongest statement is the prompt library: more than 1,400 custom prompts with prebuilt scripts for different practice areas, which suggests the product ships with qualification logic for specific matter types rather than leaving a firm to write it. No practice area is named, so a firm cannot tell whether its own is scripted. No jurisdiction is stated, and firm coverage is described only as law firms of all sizes, which excludes nobody. A number without a list describes scale, not coverage.
Practice areas are named and nothing bounds them. Legal coverage names four practice areas: personal injury, family law, estate planning and immigration. Coverage extends well beyond legal into healthcare patient intake, financial advisory and wealth management, insurance first notice of loss and quoting, internal request management, education and real estate, with use case material for each. Firm coverage is not described. There is no firm size, no jurisdiction and no statement of who the product is not for, and nothing says whether the legal configuration differs materially from the healthcare or insurance one or is the same engine pointed at different criteria. The breadth cuts both ways: a platform serving seven verticals is unlikely to carry deep legal knowledge in any one of them, and nothing published shows what the legal version knows that the generic one does not.
The 12 legal signals, side by side
Recorded rather than graded. These are the questions a practitioner has to answer before a tool touches a client matter, and the answers are taken from public material only.
Client Data in Training
Can material a lawyer puts into this product be used to train a model?
Silent. The quoted description is the product's core function as stated across independent directories, and it describes what the system does rather than what happens to what it collects. No statement in either direction was located on whether chat transcripts, text message threads, lead records or qualification outcomes are used to train or improve models. The platform holds this material durably by design, since it is a CRM as well as a chat widget, and two way business texting adds a communications archive alongside the transcripts.
No privacy policy or data processing statement was located in the material read. The people in these conversations are prospective clients who are not the customer and cannot consent, object or ask. Recorded as silent, not as a negative commitment. Checked the vendor material, the blog, the directory listings and independent review on 29 Aug 2026.
Silent, and the quoted design goal is why the silence carries weight. The vendor's stated proposition is that forms lose the context that matters and that its conversations capture the reasoning behind an inquiry, so the platform is deliberately built to elicit fuller and more personal disclosure than a form would. No statement in either direction was located on whether those conversations, the structured data extracted from them, or routing outcomes are used to train or improve models, and no retention or deletion position exists.
The horizontal footprint widens the exposure: the same undisclosed posture covers legal matter details, healthcare presenting concerns and care needs, and financial position captured in wealth management qualification. The people in these conversations are prospective clients and patients who are not the customer and cannot consent, object or ask. Recorded as silent, not as a negative commitment. Checked the home page, the Intelligent Intake page, the blog material and the site navigation on 29 Aug 2026.
Prompt and Output Retention
How long does the product keep what a lawyer typed, and can that be set to zero?
Chat transcripts and lead records persist for no stated period. No retention period is published for chat transcripts, text message history, lead records or automated follow up sequences, and no deletion right is described. Retention is structural rather than incidental: the platform provides lead tracking, customer history and interaction tracking as named capabilities, all of which require durable storage, and follow up automation depends on a lead record persisting after the conversation ends.
Nothing states what happens to records of prospective clients who never became clients, or to the database on termination. Checked the vendor material, the directory listings and independent review on 29 Aug 2026.
Intake conversations and their summaries persist for no stated period. No retention period is published for intake conversations, extracted structured data, generated summaries or routing decisions, and no deletion right is described. Retention is implied by the product's own workflow, since structured summaries are routed onward to an inbox, CRM or Slack while the underlying conversation is the only place the extraction can be checked against, and nothing states whether the vendor retains that conversation, for how long, or whether a firm can purge records of people who never became clients. Checked the home page, the Intelligent Intake page and the blog material on 29 Aug 2026.
Ethical Walls and Matter Segregation
Does retrieval respect the firm’s ethical walls, or can the model read across them?
Lead records carry no described access limits or segregation. No permission model, access restriction or segregation description was located. The chat captures a person's account of their legal problem and creates a lead record before any conflicts check, and nothing says whether that record is held back, restricted within the firm, or immediately visible to whoever works the queue. No conflict checking feature is named, and nothing addresses segregation between customers. Checked the vendor material, the directory listings and independent review on 29 Aug 2026.
Separation is claimed and not documented, though the claim is well shaped. The vendor states that the platform captures the parties and matter details needed to flag conflicts during the intake conversation, then routes the structured summary to staff who confirm against the firm's own records. That is the right division of labor: the machine gathers the facts a conflicts check needs, and a person performs the check against the authoritative record, rather than the system asserting a determination it cannot make.
Nothing further is documented. No statement covers what happens to intake data between capture and confirmation, whether it is held back pending the check or visible across the firm at once, and no permission or segregation model is described within a firm or between customers.
Third Party Request and Subpoena Notice
If someone subpoenas the vendor for a firm’s data, does the firm hear about it first?
Prospective clients' messages carry no notice protection against government requests. No government or law enforcement request clause, no commitment to notify a customer before producing their data, and no transparency report were located. The vendor holds chat transcripts and two way text message threads in which prospective clients describe legal problems, and the messaging path traverses an unidentified carrier layer, so a request could reach either the vendor or a processor the firm cannot name.
Nothing published addresses any of it. Checked the vendor material, the directory listings and the site navigation on 29 Aug 2026.
Requests for intake data from any direction meet no stated process. No government or law enforcement request clause, no commitment to notify a customer before producing their data, and no transparency report was located. The material held is broad because of the horizontal footprint: legal matter details in immigration and family law, healthcare presenting concerns and financial position are all captured through the same platform under a single undisclosed posture.
Nothing published addresses requests to the vendor from any direction. Checked the home page, the Intelligent Intake page, the blog material and the site navigation on 29 Aug 2026.
Primary Law Corpus Provenance
Where does the law in this product come from, and does the vendor have the right to use it?
Corpus provenance largely falls away here, with one live residue. The platform has no primary law corpus: the conversational content is the vendor's library of more than 1,400 authored prompts plus whatever the firm configures, so the substance is authored rather than retrieved. The residue is what the AI layer was built on. Natural language processing and language detection are stated capabilities and both are learned behaviors, and nothing published states what corpus supports either, whether the prompt library was derived from prior customer conversations, or whether qualification logic reflects aggregated intake data across firms. Checked the vendor material, the blog, the directory listings and independent review on 29 Aug 2026.
Firm defined criteria drive the content, leaving a live residue about what the models learned from. The platform has no primary law corpus: intake conversations run against criteria the firm defines, so the substantive content is customer supplied. The residue is what the conversational and extraction models were built on. Conducting an adaptive interview that follows up on vague answers and mapping free text into a schema are both learned behaviors, and nothing published states what corpus supports either, whether any of it is domain specific to legal, healthcare or insurance, or whether prior customer conversations contribute.
The vendor states that configuring an agent requires no script, which means the conversational competence comes from the model rather than from customer authored content, making the provenance question more rather than less material. Checked the Intelligent Intake page, the home page and the blog on 29 Aug 2026.
Good Law Verification
Does the product tell you when the authority it just cited has been overruled?
Citation checking does not apply here. Intaker runs intake chat, manages leads and sends follow up messages. It produces no legal research or citation to authority, so there is nothing for a citator to check. Checked the vendor material and the directory listings on 29 Aug 2026.
Intake qualification gives a citator nothing to check. Perspective AI holds intake conversations, extracts structured data and routes summaries. It produces no legal research or citation to authority, so there is nothing for a citator to check. This is a fact about the product's scope rather than a disclosure failure. The product's legal role ends at qualifying an inquiry and handing a summary to staff, before any research or advice begins. Checked the Intelligent Intake page, the home page and the legal use case material on 29 Aug 2026.
Refusal and Uncertainty Behavior
What does the product do when the answer is not in the corpus?
Uncertain moments have no described handling, and the scripted design makes the gap specific. A chat running on more than 1,400 written prompts will meet inquiries that match none of them, and nothing says what happens then. The system might say it cannot help, offer a human, give a generic response, or pick the nearest prompt and answer something next to the question. Language detection adds to it, since a misdetected language produces a reply in the wrong language, and nothing says whether low confidence detection is handled differently.
Live Call Connect offers a route out of a stuck exchange through a phone transfer, but nothing says whether the system offers it when it cannot proceed or only when the prospective client asks.
Unresolved fields have no described handling, and the extraction layer is where it matters. The vendor states that the system follows up on vague answers, which is a real uncertainty behavior in the conversation and the closest thing published. The harder case is not described: what happens when follow up fails and a required field still has no supportable value. Nothing states whether the field is left empty, filled with a best guess, or flagged as uncertain in the summary a firm acts on.
The vendor markets automation triggered by validated structured outputs, so an unflagged guess would flow into routing without a person seeing the conversation it came from. Nothing says whether the agent will decline to qualify, or escalate a case it cannot resolve, rather than routing it as a weak fit. Checked the home page, the Intelligent Intake page and the blog material on 29 Aug 2026.
Fabricated Citation Record
Does a public court record exist addressing fabricated or hallucinated legal citations in output from this product?
None located, with the instrument named. General web searches combining the vendor and product names with court, order, sanction and complaint terms returned nothing on 29 Aug 2026, and no named docket database, court record tracker or state consumer protection register was searched. Recorded as a statement about what this search found, not as a clearance. The exposure shape is not fabricated citations, since no legal authority is generated: the analogous adverse findings would be a complaint arising from an automated intake exchange, or a dispute over automated text messages sent to prospective clients under messaging and consumer protection regulation, and neither would surface through a citation focused search.
None located, with the instrument named. General web searches combining the vendor and product names with court, order, sanction and complaint terms returned nothing on 29 Aug 2026, and no named docket database, court record tracker or state consumer protection register was searched. Recorded as a statement about what this search found, not as a clearance. The exposure shape is not fabricated citations, since no legal authority is generated: the analogous adverse finding would be a complaint arising from an automated intake conversation, a misextracted fact that caused a matter to be misrouted or missed, or a disclosure issue in the healthcare configuration, none of which a citation focused search would surface.
Bar Guidance Alignment
Has the vendor engaged in public with the ethics opinions its buyers are bound by?
Professional conduct guidance has no presence in the product's account of itself. No ethics opinion, ABA Formal Opinion 512, state bar guidance or engagement with professional conduct rules was located, although the vendor publishes a blog for law firms comparing chat options and advising on intake practice. Two features engage professional rules directly and neither is addressed. Automated email and text follow up to prospective clients falls under lawyer advertising and solicitation rules that vary by state.
The video feature presents a named attorney inside an automated conversation, which engages rules on communications about a lawyer's services. Checked the blog, the vendor material, the directory listings and the site navigation on 29 Aug 2026.
Rules on communications with prospective clients go unaddressed. No named ethics opinion, ABA Formal Opinion 512, state bar guidance or engagement with professional conduct rules was located. The vendor publishes a substantial analytical library on conversational intake, including a comparison of legal intake platforms and buyer frameworks aimed at firms. It addresses the legal market analytically and reaches none of the rules governing communications with prospective clients, automated advice or advertising.
Those rules govern exactly the conversations the product holds. Checked the blog index, the legal intake comparison material, the home page and the Intelligent Intake page on 29 Aug 2026.
Billing and Fee Posture
Does the vendor address what happens to the bill when the work takes an hour instead of six?
The product works before an engagement exists, so it sits outside the fee relationship, and its savings claims are framed as revenue rather than cost. The vendor says its workflow sequences and automation tools help law firms maximize revenue, and its positioning is signing more cases with less effort and better lead conversion. Independent directory material notes that buyers judge total cost on lead conversion and automation depth rather than subscription price.
No figure, method or baseline supports those revenue claims, and no customer outcome number was located. Nothing addresses whether an automated intake exchange or follow up sequence is disclosed to the prospective client as machine generated, and no record shows which part of a first interaction was automated. Checked the vendor material, the directory listings and independent review on 29 Aug 2026.
The product operates before an engagement exists, and no savings claim is published. No time saving, conversion improvement or return on investment figure attributable to this vendor's customers was located. The vendor does publish industry figures in its analytical content, including form completion rates of 20 to 40 percent and deflection and cycle time improvements attributed to insurance carriers generally, but those are category research rather than product outcomes.
Nothing appears on the client's side. There is no position on whether an automated intake interaction is disclosed to the person as handled by a machine, and no record showing which part of an intake was automated. Checked the home page, the Intelligent Intake page and the blog material on 29 Aug 2026.
Outside Counsel Guideline Readiness
Can a firm get this vendor through a client’s AI clause without a bespoke negotiation?
Vendor paperwork for client disclosure is absent. No trust center, security page, named certification, subprocessor list, named model provider, data processing agreement or documentation request route was located, so a firm has nothing to forward to a client. The gap goes beyond the model. A firm asked which third parties handle its prospective clients' text messages and phone transfers could not answer, because neither the messaging carrier nor the telephony provider behind Live Call Connect is named. Checked the vendor material, the blog, the directory listings and the site navigation on 29 Aug 2026.
Neither firms nor medical practices get any disclosure material to forward. No trust center, security page, named certification, subprocessor list, named model provider, data processing agreement or route to request documentation was located, so a firm has nothing to forward and nowhere to point a client. The gap reaches beyond legal. The same platform is marketed for healthcare patient intake with no HIPAA posture or business associate agreement published, so a medical practice would face the same problem. Checked the home page, the Intelligent Intake page, the blog material and the site navigation on 29 Aug 2026.
Court Disclosure Support
If a judge’s standing order requires an AI disclosure, can the product produce one?
A producible record of each conversation is missing, though the raw material exists. Nothing indicates that output records which model produced it, no human verification record is captured, and no export or audit record was located. The platform keeps a lot of raw material by design: customer history and interaction tracking are named capabilities, and chat transcripts and text threads sit in the CRM, so evidence of what was said exists in principle.
Nothing states that it can be produced as a record or that it separates automated responses from human ones. That distinction matters here, because one conversation can contain automated prompts, a recorded attorney video and a live call, and nothing describes a record showing which was which. The likely forum is a bar complaint, a fee dispute or a malpractice claim rather than a court filing.
The routed summary is the lasting record, and the conversation behind it has no stated home. Nothing indicates that output records which model produced it, no human verification record is captured, and no export or audit artifact was located. The product does create a lasting artifact in the ordinary course: the structured summary routed to an inbox, CRM or Slack. That is a record of what the system concluded, not of what was said, and nothing states whether the conversation transcript is kept alongside it or available to the firm.
That distinction is the whole question. If a prospective client later disputes what an intake agent asked or told them, or a firm needs to show why a matter was routed as it was, the summary is the machine's account and the conversation is the evidence. The forum here is a bar complaint, a fee dispute or a malpractice claim rather than a filing. Checked the home page, the Intelligent Intake page and the blog material on 29 Aug 2026.
The questions both sides leave open
Derived from the records above rather than written, so it cannot favor either vendor. Take these into both conversations and ask each side the same question.
- Citation Accuracy and Hallucination Disclosure
- Privilege and Confidentiality Posture
- UPL and Professional Responsibility Posture
- AI Governance and Bias Disclosure
- AI Safety and Data Stewardship
- AI Liability and Recourse
- Deployment Model and Data Residency
- Security Certifications and Trust Center
- Model Supply Chain Disclosure
- Prompt and Output Retention
- Third Party Request and Subpoena Notice
- Primary Law Corpus Provenance
- Good Law Verification
- Refusal and Uncertainty Behavior
- Bar Guidance Alignment
- Outside Counsel Guideline Readiness
- Court Disclosure Support
Which one fits
Choose Intaker if
- You want the prospect on the phone while they are still interested. Intaker's Live Call Connect turns a chat lead into a real time phone transfer to the firm through a click to call widget. That moves the lead from machine to human inside the same interaction, rather than handing over a record afterwards. Attorney recorded intro and outro videos play in the chat, so a named person is visibly present in an automated conversation.
- The lead has to land in the system you already run. Intaker names Clio and Filevine on the practice and case management side, alongside Salesforce, Calendly, Google Analytics and Google My Business. The Google My Business connection suits consumer legal work, where a large share of inquiries begin at a local listing rather than at the firm's own website.
- You want qualification logic you do not have to write. Intaker states more than 1,400 custom prompts with pre built scripts for different practice areas, so matter specific qualification ships with the product. Around it sit two way business texting, automated follow up sequences and internal workflow reminders. Independent software directories, not Intaker, report a starting price of 80 dollars a month.
Choose Perspective AI if
- Forms lose the thing you needed. Perspective AI conducts a goal directed interview instead. It asks follow up questions when an answer is vague, adapts on the responses, and captures urgency and the reasoning behind an inquiry. Its schema aligned extraction puts unstructured answers into the same fields a form would have collected, while keeping the context a form discards.
- You want to read the script before a prospective client does. Configuration starts from a description of who the firm is trying to qualify and what determines fit, and the conversation is built around it. Questions can then be adjusted, routing rules set for strong against weak fits, and branding matched. The full conversation can be previewed before it goes live.
- Conflicts cannot be decided by a machine. Perspective AI states that the platform captures the parties and matter details needed to flag conflicts during the intake conversation. It then routes the structured summary to staff, who confirm against the firm's own system of record. The system gathers and a person decides, on the one question where an automated determination would be indefensible.
In summary
Intaker
Intaker is a conversational intake and client relationship platform for law firms, built around website chat that pre qualifies leads around the clock. The chat runs on more than 1,400 custom prompts with pre built scripts by practice area, and attorney recorded video plays inside it. Live Call Connect transfers a chat lead to a phone call in real time, and a CRM covers lead tracking, two way texting and automated follow up. According to the AI Legal Index, Intaker names Clio, Filevine, Salesforce, Calendly and Google My Business as integrations. It publishes no security attestation, data handling position, model provider or price.
Perspective AI
Perspective AI is a conversational intake platform that replaces static forms with adaptive AI led interviews. It asks follow up questions on vague answers and captures urgency and the reasoning behind an inquiry. Its schema aligned extraction puts unstructured answers into the fields a form would have collected. It is sold across several verticals, with a named legal use case covering personal injury, family law, estate planning and immigration. According to the AI Legal Index, the model is the entire product: remove it and nothing remains, because the thing being replaced is the form. Firms can preview the full conversation before deployment, and conflicts detail is routed to staff to confirm. It names no customer and publishes no security attestation or data handling position.
Questions buyers ask
Intaker vs Perspective AI: which is better for intake conversion?
They are built for different jobs. In Perspective AI the model is the product, and removing it leaves nothing. Intaker's strength is where the lead lands, naming Clio and Filevine alongside Salesforce and Google My Business, and offering a live transfer from chat to a phone call. From the AI Legal Index, based on each vendor's own published materials as of September 3, 2026. No vendor pays for placement.
Scripted or generated, and what is the difference in practice?
Intaker runs on more than 1,400 authored prompts and pre built scripts by practice area, so the conversation is written in advance and the qualification logic ships with the product. Perspective AI generates the conversation, following up on vague answers and adapting as it goes, then extracting the result into structured fields. The trade off is real in both directions: a scripted system produces less free text and therefore less room to say something wrong, and an adaptive one captures the detail a fixed script never asks for. From the AI Legal Index, based on each vendor's own published materials as of September 3, 2026. No vendor pays for placement.
Where does the captured intake go?
Intaker names the destinations: Clio and Filevine for practice and case management, Salesforce, Calendly, Google Analytics and Google My Business, though no integration depth is described and nothing states whether data flows one way or both. Perspective AI describes its destinations at category level, saying structured summaries route to an inbox, a CRM or Slack, with Slack the only product named, so a firm cannot confirm from published material that its own system is supported. From the AI Legal Index, based on each vendor's own published materials as of September 3, 2026. No vendor pays for placement.
Does either handle conflicts?
Perspective AI addresses it and Intaker does not. Perspective AI states that the parties and matter details needed to flag a conflict are captured during the conversation and the structured summary is routed to staff who confirm against the firm's own system of record, which puts the machine on gathering and the human on deciding. Intaker publishes nothing on conflicts, prospective client confidences or the status of a chat transcript from someone who never becomes a client. From the AI Legal Index, based on each vendor's own published materials as of September 3, 2026. No vendor pays for placement.
What do Intaker and Perspective AI both leave unpublished?
Neither publishes a security attestation of any kind, a data handling position, a retention period or a deletion route, which matters because both hold accounts that members of the public give about their legal problems. Neither names a model or a provider. Neither publishes a liability position, so nothing states who bears the loss when a qualified inquiry is wrongly routed away. And neither publishes an AI governance position or any evaluation of how qualification outcomes fall across the people being assessed. From the AI Legal Index, based on each vendor's own published materials as of September 3, 2026. No vendor pays for placement.
Neither vendor paid for inclusion, placement or a grade, and neither reviewed this page before it published. Everything on it comes from public material on the dates shown. How the index grades.
What weighs most here is what neither vendor publishes. Neither publishes a security attestation, a privacy or data handling position, a retention period, a named model provider or a liability position. Two specifics belong alongside that. Perspective AI names no customer anywhere. Its published evidence is self authored, including comparative rankings of the intake category in which it places itself first, which is marketing rather than evidence. Intaker's 80 dollar entry price comes from independent software directories rather than from a vendor pricing page, and no tier structure or unit of charge accompanies it. Neither vendor reviewed this page.