Draftwise vs Spellbook: how they compare in 2026
Draftwise and Spellbook both sell AI drafting to transactional lawyers, and both live inside Microsoft Word. The difference is disclosure about the AI itself. Spellbook names its models on its own home page, GPT-5 and Claude Opus, and names OpenAI and Anthropic as the providers. It states the zero data retention agreements negotiated with each and what they mean technically, names AWS as its cloud with Canada and the United States as processing locations, and publishes a full subprocessor list. Draftwise publishes no model, no provider, no subprocessor list and no architecture beyond the phrase agentic AI. It answers on the boundary rather than the model. It offers a private cloud in the customer's choice of AWS, Azure or GCP, where the firm's own IT holds the encryption keys and no data leaves the VPC by default, alongside on premise installation.
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.
Draftwise uses a firm's deal history to draft, review and negotiate agreements in Word. Its agent interprets redlines and comments, follows version history and applies precedent to new drafting tasks. Playbook Studio creates rules from example agreements and representation instructions, while Deal Table extracts terms, risks and obligations across the contract corpus. Ontonomi connects clients, matters, contractual relationships and positions into a structured model whose answers trace back to source material. Knowledge Console controls the collections and guidance available to these workflows. The underlying document management system remains the firm's existing repository; Draftwise supplies the intelligence that turns that material into proposed legal work.
The models are the product. Review, Draft, Ask, Compare and the Associate agent are all generative capabilities delivered through a Word add in that exists to carry them. The company pivoted its whole business to generative AI in 2022, rebranding from Rally to Spellbook. Remove the models and there is no product, only an empty add in.
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.
Ontonomi resolves questions through a structured map of the firm's legal relationships and shows the exact memo, clause or agreement behind each answer. This supplies an inspectable trail through the material used for the result. The drafting workflow also uses selected precedent from the document management system, and negotiation comparisons can draw on EDGAR filings. Playbook Studio lets a reviewer inspect citations while editing and exporting the marked agreement. Accuracy and repeatability are asserted without a reproducible test set, error rate or account of unsupported questions. Those assertions should be distinguished from the source links, which establish where an answer came from rather than whether it is correct.
Ask can use selected iManage files and return citations to the team's own contracts. The legal data connection retrieves material from named open legal databases such as CanLII, CourtListener, Cornell Law, EUR Lex and regional legal information institutes, with links to the underlying authority. Those sources can inform questions, drafting and review inside the same workflow. Their currency follows the source operator's updates. Spellbook publishes no reproducible accuracy evaluation for the current product, no automatic subsequent treatment check and no abstention threshold. A citation makes an answer inspectable, but checking the quotation or legal proposition against it is still the reviewing lawyer's job.
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.
Playbook Studio turns uploaded example contracts and representation instructions into proposed rules that the user can refine and save. Running a review produces redlines, with a surface for inspecting citations, making manual edits and exporting the marked agreement. Knowledge Console lets administrators control collections, user access and approved playbooks across the organization. These functions give lawyers and knowledge teams specific ways to influence and inspect the work. The drafting agent also performs work across deal history and whole agreements, but Draftwise does not describe its stop conditions or error escalation route. There is no complete account of which wider agent actions require separate approval.
Review proposes changes in Word for the lawyer to inspect and approve, either individually or in a group. Playbooks encode the team's instructions, while Associate takes on work across several files with user oversight. Proposed edits can carry the reviewer's identity when accepted, and citation links let the reviewer inspect supporting material. The terms place responsibility for reliance and verification with the customer. The existing suite therefore has recognizable human review surfaces, but no universal threshold is stated at which Associate must stop or escalate an unsupported conclusion. The separately announced Autonomous Contract Management offering remains in early access.
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.
Orrick, Gunderson Dettmer, Mishcon de Reya, Mayne Wetherell and Sekri Valentin Zerrouk provide named accounts of using Draftwise. Orrick partner Zac Padgett describes extracting a market percentage from comparable deal agreements during final negotiations, while Mayne Wetherell's Cameron Reeves describes quicker access to curated precedent. These examples identify practitioners and the work performed without a dated measurement method. Playbook Studio separately claims first pass creation falls from more than six hours to five minutes and NDA review from an hour to two minutes. Those figures have no named customer, sample or evaluation method attached to the product page.
Named customer accounts involve Dropbox, Panasonic, KMSC Law, Alturas Capital Partners and Elevare Law. Panasonic's account concerns saving three weeks while building an RFP process, while other testimonials describe drafting capacity and keeping work inside the legal team. The current pricing page states use by more than 5,000 legal teams and includes those stories alongside separate firm and in house offerings. These examples establish identifiable use rather than only a logo wall. A consistent dated measurement protocol, comparison baseline and workload definition do not accompany the range of savings claims, so the figures are specific customer accounts rather than a reproducible general result.
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.
Customer data does not train public models, and Ontonomi's customer owned model is not used to train a tool sold to another firm. The data policy also prohibits sharing firm intellectual property, including trained intelligence or aggregates. Live DMS permissions govern access, while private cloud and on premises options can leave encryption keys with the firm's IT team. These commitments support confidentiality but do not explain every model provider's access or retention. A customer subscription agreement is not available alongside the website terms. Privilege and work product protections are not expressly allocated, and private learning within the customer's own environment is distinct from the prohibition on shared training.
Spellbook's agreement expressly protects privileged information and work product and requires confidentiality for staff and service providers handling customer content. Its iManage integration preserves existing file permissions and ethical walls. Third party models cannot train on customer content and operate under zero retention commitments. Spellbook itself may use deidentified content for customer specific personalization and retain inputs and outputs temporarily for quality assurance, while the terms restrict those uses and access. These permissions are narrower than unrestricted cross customer training, but they are not an across the board ban on learning from client content. Application history also has no single stated in use retention deadline.
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.
The product is aimed at transactional lawyers rather than a consumer seeking personal legal advice. It drafts agreements, applies precedent and analyzes negotiation positions, with lawyers able to inspect citations and revise proposed changes. Its ethical wall guidance discusses client restrictions, access control and material that should never reach AI. Draftwise makes no explicit statement separating software assistance from legal advice and sets out no competence or supervision requirements across the jurisdictions served. The website terms disclaim accuracy in general commercial language. Neither those terms nor the professional audience amounts to a jurisdiction specific framework for responsibility when generated contract language is wrong.
Output counts as information rather than formal legal advice under Spellbook's terms, which place verification and appropriate lawyer review with the customer. The product is directed to law firms and legal departments, with review controls inside Word and oversight of Associate's work across files. The agreement also warns that non English use can produce unsatisfactory results. That is a concrete language limitation, but not a tested jurisdictional coverage matrix or a map of which regulated services the software can provide. Human acceptance of a redline remains part of contract work, and an answer from a cited public source is not a substitute for checking its legal application.
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.
Knowledge Console organizes access to precedent, users, licenses and approved playbooks, while usage analytics show adoption and engagement. The ethical wall guidance adds live permission checks and restrictions on material that may enter an AI workflow. These are controls over a firm's knowledge and access, not a release assurance framework for the vendor's models. Draftwise names no model risk owner, describes no evaluation procedure before deployment and gives no results across legal practices or populations. Information security policies assign responsibilities to the executive team and require regular review, but do not explain how incorrect legal answers are investigated or how model behavior changes are approved.
Spellbook's terms describe internal AI risk and prohibited use policies, staff education and regular reassessment of ethical and legal requirements. Its security material also cites a CMS Law opinion classifying the product under the EU AI Act and identifying applicable controls. These are concrete governance arrangements beyond a general promise to act responsibly. The product supports verification through citations and user approval of edits, but those controls are separate from evaluating a model before release. The framework names no internal owner, no reproducible release test set and no results on uneven behavior across populations or matter types.
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.
Encryption uses AES 256 at rest and TLS 1.2 or 1.3 for transport. CloudWatch collects operational logs with alerting and anomaly detection, retaining them for up to two years. The separate retention policy sets a general maximum of five years, subject to legal requirements, and deletes data when its applicable period ends. Playbook Studio promises to delete trial uploads after the trial finishes. Information security rules restrict employee access by business need and require escalation of unusually sensitive material. Draftwise publishes no named processor register, exact AI prompt lifecycle or customer breach notification timetable, and the general privacy page's retention heading does not supply those periods.
OpenAI and Anthropic process Spellbook requests under zero retention and no training commitments. Spellbook's own terms permit quality assurance copies of inputs and outputs for up to 90 days, restricted to authorized staff, and self service abuse monitoring for up to 30 days unless an investigation or law requires longer. Microsoft identity integration lets organizations enforce their authentication policies, and iManage access preserves source permissions. The agreement covers confidentiality, security incidents and deletion or return of customer material, while the processor register identifies downstream services. Outside those secondary uses, Spellbook gives no single fixed period for how long application history is kept.
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 August 2025 website terms address access to the site and connected media rather than setting out a complete enterprise subscription agreement. They disclaim warranties of accuracy and completeness, exclude liability for direct and indirect damages arising from site use, and require the user to indemnify Draftwise for specified claims. They also disclaim responsibility for loss or corruption of transmitted data. No monetary cap appears in those clauses. Those clauses contain no AI output warranty, customer indemnity or contractual remedy for a failed legal workflow. A cyber insurance certificate is available by request; its cover and exclusions cannot be determined from the resource title.
Specified intellectual property claims against the software are covered by an indemnity, alongside service warranties, while the accuracy of generated output and responsibility for customer reliance on it are disclaimed. The general limit is twelve months of fees, with exceptions including confidentiality, security, data incidents and indemnification duties. The agreement includes mutual protection for specified data incident claims and gross misconduct. Beta Features have separate warranty exclusions and a CAD 10,000 cap. These are concrete protections for platform and data risks, but they do not provide an accuracy warranty compensating a client when an accepted clause or legal assertion is wrong.
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.
Word is the drafting surface, and the connected DMS supplies agreements, clauses, tags and version history. Deal Table brings together material from iManage, NetDocuments, Google Drive and SharePoint to analyze terms and obligations. Ontonomi exposes its structured knowledge through MCP and APIs. The access approach checks the DMS using the individual lawyer's identity and live permissions, with access events intended to remain visible in the source system's audit trail. The product pages describe what those connections support, but do not provide an implementer's complete setup guide, synchronization schedule or field mapping. Customer cloud deployment also requires coordination with the firm's IT environment.
The iManage connection starts from Add from iManage in Ask or an Associate Project and authenticates the user's account once. The same connection then works in Word and Associate. Users select files through the attachment control or project file list; selected content becomes available as chat context or project references without a manual download. Answers cite the imported contracts, and precedents can guide review of incoming paper. Existing iManage permissions and ethical walls govern file access. The connection describes movement from iManage into Spellbook's working context; it does not promise automatic return of revised files to the source repository.
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.
A firm can use a dedicated private cloud on AWS, Azure or Google Cloud, deploy into its existing cloud, or run Draftwise on premises. For the private cloud option, the firm's IT controls encryption keys and auditing, vendor access is limited to support, and network access can be restricted to firm subnets. Existing cloud installation can use containers, virtual machine images or Linux binaries, with administration by Draftwise or the customer. These are concrete hosting and tenancy choices. The default statement that data stays inside the VPC does not separately identify the location of every language model inference service, so storage control is clearer than the complete AI processing geography.
Customer content is stored and processed in Canada and the United States, Spellbook states, with AWS as its primary cloud provider. Third party model requests are handled under zero retention agreements with OpenAI and Anthropic, while Spellbook itself has separate retention rights for quality and abuse monitoring. The Word add in and Associate connect to shared cloud services, including optional sources such as iManage. The public description identifies storage and processing countries but does not offer a per tier residency selector or a dedicated customer deployment. A source system's location therefore does not confine every later processing step to that same 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.
Draftwise's trust center lists SOC 2 Type I, SOC 2 Type II and ISO 27001:2022, with an ISO certificate, a SOC 2 report and a penetration testing attestation dated January 22, 2026 among its resources. Access to the certificate and report requires a request, so their auditor, scope and coverage period are not visible from the listing alone. The FAQ additionally names Cyber Essentials Plus. These standards concern the company's security rather than the accuracy of generated legal work. The portal also lists a 2026 cyber insurance certificate and disaster recovery testing material, with the underlying terms subject to access.
Spellbook states that it is SOC 2 Type II compliant and provides a trust portal for reports, processor information and compliance resources. Its security program uses encryption, Microsoft identity controls and agreements governing model providers. The public security explanation also describes HIPAA support through business associate agreements where appropriate and a legal opinion on EU AI Act classification. These are different assurances with different scopes. The open product explanation leaves out the current SOC audit period, the auditor and the full service scope. The named SOC standard therefore supports a specific certification claim without treating every resource or regulatory statement as an equivalent independent security audit.
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.
Ontonomi supplies a structured legal knowledge model over the firm's clients, matters, agreements and negotiated positions. It connects those relationships and traces answers back to their source clauses or memoranda, while the drafting system applies precedent and review rules to Word agreements. The knowledge architecture supporting the AI workflows is therefore described. Draftwise does not name its language model providers or versions, or say how individual tasks are allocated among models. AWS, Azure and Google Cloud are hosting choices rather than an identification of inference suppliers. The promise against training public models leaves open where each provider processes a request and how substitutions are communicated.
Language models come from OpenAI and Anthropic, under zero retention agreements that keep request and response content in memory for processing rather than persistent storage at those providers. AWS is the primary cloud provider, and storage and processing are described in Canada and the United States. The trust portal lists processors and records updates when suppliers change. Spellbook does not give the exact production model versions or a map of which model serves each feature. A public processor update history is not an explicit commitment to notify every customer whenever a model version or deployment location changes within the same supplier.
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.
Draftwise requires a sales discussion for product pricing. Neither the core product pages nor the FAQ give a rate, charging unit, tier structure, deployment fee or enterprise minimum. Playbook Studio has a free trial registration route and promises deletion of uploaded material when that trial ends, without stating an ongoing subscription price there. The website terms describe card, wire and ACH payments, recurring billing and cancellation at the end of the paid term. Those payment mechanics say nothing about the cost of a firm deployment. Private cloud, existing infrastructure and on premises arrangements also lack a stated implementation charge.
The number of team members on a license sets the price of the suite. Law firm and in house packages both include the Word add in and Associate, training and online support, with dedicated support for teams above ten. The in house offering includes a playbook build service. A seven day trial is available, with extended trials possible for larger organizations. Subscription amounts remain custom rather than a public per user rate. The terms describe annual renewal by default and included maintenance and support, but the price of a particular deployment and any separately agreed services remains part of its order.
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.
Large and midsize law firms and corporate legal teams use Draftwise for transactional work. Its in house examples span advertising technology, defense contracting, investment funds, life sciences, real estate and software companies. Ontonomi adds concrete relationship models for funds and side letters, acquisitions, lending arrangements and corporate contract obligations. Named customers include Orrick, Gunderson Dettmer, Mishcon de Reya, Mayne Wetherell and Sekri Valentin Zerrouk, reflecting practice across several legal markets. The depth of precedent available depends on the firm's own connected corpus. Draftwise does not list unsupported agreement types, set jurisdiction boundaries or say how far support extends beyond transactional practice.
Law firm and in house packages are sold separately, and the principal tools center on transactional drafting, review, negotiation and questions about agreements. Associate extends work across several files, and iManage access allows firms to use existing contract precedents. Industry material covers energy, healthcare, finance, technology, manufacturing and consumer businesses. The legal data connection adds government and court sources across many regions, though breadth of sources is a different matter from tested competence in every practice. The terms warn about non English results, while Autonomous Contract Management remains in early access. Spellbook does not list the jurisdictions or specialized legal tasks it leaves unsupported.
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?
Customer material is excluded from training public models, and Ontonomi says the firm's data never trains a tool sold to another firm. The customer owns its ontology and the information within it. The FAQ also prohibits disclosure of firm intellectual property, including trained intelligence and aggregates. These are product policy commitments rather than an express training clause in a customer subscription agreement. They do not prohibit learning confined to the firm's own environment or identify every external model provider's terms.
The terms prohibit third party models from retaining or training on customer content. Spellbook itself may process deidentified customer material to improve that customer's features and calibrate personalized systems, without using that personalization for other customers. It can also evaluate prompt and system improvements through limited quality monitoring. These own service improvement permissions are distinct from the provider prohibition and do not require a separate opt in in the standard agreement.
Prompt and Output Retention
How long does the product keep what a lawyer typed, and can that be set to zero?
A general retention policy sets a maximum of five years unless legal or regulatory requirements require longer, with the actual duration varying by dataset and purpose. Operational logs can remain for up to two years. Playbook Studio trial uploads are deleted when the trial ends, and the security program includes data deletion when customers leave. Draftwise gives no precise period for production prompts, generated drafts or residual backups. Customer hosted deployment also puts some storage under the firm's administration.
Model providers handle request and response content in memory under zero retention agreements. Spellbook may keep quality assurance inputs and outputs for up to 90 days, and self service abuse monitoring data for up to 30 days, with investigation and legal exceptions. Those are specific secondary uses rather than a stated ordinary lifetime for all application history. Deletion and return provisions are a different mechanism from a configurable zero history setting.
Ethical Walls and Matter Segregation
Does retrieval respect the firm’s ethical walls, or can the model read across them?
Ontonomi enforces identity across connections and mirrors DMS permissions live. Draftwise's ethical wall guidance describes checking access as the individual lawyer at the time of each request, rather than relying on permissions copied at ingestion. It also describes blocking selected client, matter, document or clause material from AI and preserving access events in the source DMS audit trail. Knowledge Console controls collections and users, but the source repository remains the authority for whether a person can see a particular file.
The iManage integration applies existing iManage permissions and ethical walls to file access inside Spellbook. A user authenticates the source account once, then selects permitted files for Ask in Word, Ask in Associate or an Associate Project. Selected material becomes context for answers and review. Spellbook describes this protection specifically for the iManage connection and does not extend the claim to permission inheritance on the other optional storage connections.
Third Party Request and Subpoena Notice
If someone subpoenas the vendor for a firm’s data, does the firm hear about it first?
The privacy policy permits disclosure of personal data in response to lawful public authority requests, including national security and law enforcement demands. Where law permits, Draftwise limits the disclosure to what is legally required. It does not commit to telling the customer before responding or describe a request transparency report. Private deployment may affect which material the vendor possesses, but does not itself create a notification obligation. A separate negotiated customer agreement could contain additional terms not available in the website policies.
Spellbook's terms require prompt notice of a legal or government demand for customer data and reasonable cooperation with efforts to obtain a protective order or contest disclosure. The confidentiality provisions also address compelled disclosure and protection of privileged material. These contractual steps give the customer an opportunity to respond to a demand. Spellbook does not publish a report counting government requests and their outcomes.
Primary Law Corpus Provenance
Where does the law in this product come from, and does the vendor have the right to use it?
Draftwise draws principally on a firm's agreements, clauses, tags and deal history, with EDGAR filings used for market comparisons. Ontonomi links those materials into a map of relationships and negotiated positions. Neither customer precedent nor filed commercial agreements identifies a collection of primary case law or legislation behind legal research answers. Draftwise names no primary law corpus, licensing basis or update schedule. The cited source trail serves contract knowledge and negotiation work rather than a law publisher collection.
The legal data connection names its sources, including CanLII, Cornell Law, CourtListener, EUR Lex, BAILII and national government registers. Spellbook retrieves material directly from each source of record and follows its update cadence. The same workflow can use the customer's own contracts and iManage precedents. The named source list makes jurisdiction coverage inspectable, but does not state an individual license or public domain basis for every external collection or its separate market benchmark data.
Good Law Verification
Does the product tell you when the authority it just cited has been overruled?
The workflow compares contractual positions and retrieves precedent from earlier matters. EDGAR benchmarking can show how a term appears in filed agreements, and Ontonomi traces a result to the clause or memorandum behind it. Neither function establishes whether a judicial authority remains good law. Draftwise describes no citator, subsequent history method or status warning for overruled decisions. The product is primarily aimed at transactional drafting and negotiation, so its market comparisons should not be treated as legal authority checks.
Citations to public case law, statutes and registers come back from the legal data feature, with currency tied to the operator of each source. Ask, Draft and Review can use that authority within the contract workflow. The feature does not describe a licensed citator or a method for checking subsequent judicial treatment of a case. A current link to the underlying decision is therefore separate from a signal that its legal proposition remains reliable.
Refusal and Uncertainty Behavior
What does the product do when the answer is not in the corpus?
Ontonomi promises repeatable answers with a trace back to source material, while Playbook Studio lets a lawyer inspect citations and edit proposed redlines. Draftwise does not define a response for when the required support is absent. There is no stated abstention message, calibrated uncertainty display or escalation rule for a drafting request that the firm's precedent cannot support. A source trail can help a reviewer investigate an answer, but does not by itself describe what the system does when it should not answer.
Source citations and selected precedent files provide context for Spellbook's answers and proposed redlines. A lawyer can inspect that material and accept or reject suggested changes. The product does not describe an automatic no answer response, an exposed confidence threshold or a required escalation when available material cannot support a conclusion. A contractual warning about unreliable output assigns verification to the user rather than defining an abstention mechanism.
Fabricated Citation Record
Does a public court record exist addressing fabricated or hallucinated legal citations in output from this product?
No court incident naming Draftwise was identified in the AI Hallucination Cases database on October 16, 2026. Draftwise's main work concerns contractual precedent, negotiation and deal analysis rather than preparation of litigation filings. Its expanded source trails and editable review results give practitioners material to inspect, which is not the same as error free output. The incident result concerns records that name the tool and is separate from a controlled assessment of citation or drafting accuracy.
Youssian v Village of Kings Point, 2026 NY Slip Op 26150, Supreme Court of New York, Nassau County, September 9, 2026, addresses counsel's fabricated quotations and misrepresented holdings. The opinion recounts counsel accepting Spellbook AI changes without repeating a citation check. It sanctions counsel $5,000, awards $10,000 in fees and strikes both petitions. The findings concern the lawyer's verification and response to the errors; the product attribution appears in counsel's explanation.
Bar Guidance Alignment
Has the vendor engaged in public with the ethics opinions its buyers are bound by?
Draftwise's ethical wall guidance addresses the firm's responsibility to enforce client restrictions, live access permissions and limits on sending material to AI. It discusses protective orders, client guidelines and the risk of creating a second permission system outside the DMS. The article presents these as standards Draftwise follows, without mapping them to a named bar opinion or a jurisdiction's professional conduct rules.
Competence, billing and client disclosure obligations are not translated into a complete product specific compliance guide.
Spellbook's educational material addresses competence, candor and checking AI output in general terms, and the company has a relationship with the Canadian Bar Association. Its terms also require appropriate legal review and distinguish output from formal advice. Those positions concern professional responsibilities, but the available account does not map the suite to a named ethics opinion or a set of jurisdiction specific bar requirements. A partnership alone does not supply that mapping.
Billing and Fee Posture
Does the vendor address what happens to the bill when the work takes an hour instead of six?
The product emphasizes faster preparation, contract review and access to a firm's negotiated precedent. Named practitioners describe spending less time finding language and extracting deal insights, while Playbook Studio gives time reduction figures for creating and applying playbooks. Knowledge Console tracks adoption and engagement, but does not describe a client billing record that distinguishes AI work from a lawyer's review time. Draftwise does not address how saved time, subscription charges or AI assisted tasks appear on a legal invoice.
The law firm suite is pitched as a way to improve profitability and the in house suite as a way to retain more work internally. Customer stories describe time savings, including Panasonic's RFP process. The product's per user subscription is separate from a lawyer's charges to clients. A matter level invoice record of AI assisted time or instructions for disclosing those savings on a client's bill are not described in the current pricing offer.
Outside Counsel Guideline Readiness
Can a firm get this vendor through a client’s AI clause without a bespoke negotiation?
The trust center offers security reports, certification, testing and insurance material through an access request, while privacy, information security and retention policies can be read directly. Ethical wall guidance discusses client restrictions and blocking material from AI. Those materials support procurement review, but Draftwise describes no client disclosure package naming model providers and their processing locations.
A complete named subprocessor register is also absent from the accessible policy pages. The available reports require the firm's own assessment against each client's instructions.
Spellbook identifies OpenAI and Anthropic, provides a processor register and explains provider retention and training restrictions. Its security resources address SOC 2 Type II, HIPAA support and an EU AI Act opinion, while the agreement covers privileged information and legal demand notice. Those materials support a firm's client review. They do not form a single client consent pack, and the firm's account must distinguish provider restrictions from Spellbook's own personalization and monitoring permissions.
Court Disclosure Support
If a judge’s standing order requires an AI disclosure, can the product produce one?
Ontonomi ties answers to the source clauses, memoranda and agreements used in its reasoning. Playbook Studio supports citation inspection, manual editing and export of a marked agreement. The access approach also preserves DMS events under the individual user's identity, while CloudWatch logs operational activity separately. These records capture parts of source use and revision work, but do not form a complete court disclosure history identifying model versions, prompts and lawyer approval for every output. Draftwise describes no filing template or dedicated certification export.
Word review retains proposed tracked edits, and answers based on contracts or public legal data include source citations. Those features preserve portions of the review process and allow a lawyer to check supporting material. An export combining exact model version, retrieved sources and human verification for each generated passage is not described. The available editing and citation records are therefore partial support for an account of AI use rather than a complete court certification package.
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.
- Good Law Verification
- Refusal and Uncertainty Behavior
Which one fits
Choose Draftwise if
- Client mandates mean the work cannot leave your infrastructure. Draftwise publishes three deployment routes: a private cloud dedicated to the firm in its choice of AWS, Azure or GCP where the firm's own IT holds the encryption keys and audit control and no data leaves the VPC by default, installation into the firm's existing cloud through a helm chart, VM images or Linux binaries, or an on premise install.
- Your precedent is the asset, not a generic corpus. Draftwise connects to the firm's document management system, naming iManage and NetDocuments, reads the complete deal history including documents, clauses and tags, generates playbooks automatically from that precedent rather than by hand, and states that it mirrors the document management system's permissions.
- You want references who will speak on the record. Draftwise publishes attributed quotes from named executives at Orrick and Gunderson Dettmer, a managing associate at Mishcon de Reya and partners at Mayne Wetherell and Sekri Valentin Zerrouk, with dedicated case study pages for three of them. None of the quotes carries a figure.
Choose Spellbook if
- A client asks which models see its draft. Spellbook names GPT-5 and Claude Opus on its own home page, names OpenAI and Anthropic as the providers, states the zero data retention agreements negotiated with each and what they mean technically, names AWS as its cloud, gives Canada and the United States as processing locations, and publishes a full subprocessor list in its trust portal.
- You want governance evidence rather than a principles page. Spellbook commissioned an independent legal opinion from CMS Law on its classification under the EU AI Act, publishes the full opinion letter for download alongside the controls implemented for that classification, and links its SOC 2 Type II and HIPAA badges to dated resources in the same portal rather than leaving them as images.
- You want to know how the answer is grounded. Spellbook states its architecture openly: it does not fine tune, it connects general purpose models to proprietary market data and requires them to fetch and cite, on the stated reasoning that putting documents into a model's long term memory encourages hallucination, with Compare benchmarking a clause against more than 2,300 contract types.
In summary
Draftwise
Draftwise is a contract drafting, review and negotiation tool for transactional lawyers that works from a firm's own precedent rather than a generic corpus. It connects to the document management system, reads the complete deal history and operates inside Microsoft Word. According to the AI Legal Index, its case rests on deployment. It publishes a private cloud in the customer's choice of AWS, Azure or GCP, where the firm's own IT holds the encryption keys and no data leaves the VPC by default, alongside installation into a firm's existing cloud or on premise. It names no model or provider and publishes no subprocessor list, no AI governance material and no pricing at any level.
Spellbook
Spellbook is AI contract review and drafting for transactional lawyers, delivered mainly as a Microsoft Word add in and also working in Google Docs. It covers redlining against a firm's standards, drafting from precedent, playbooks, cited answers and clause benchmarking, with an agent for longer multi document matters. According to the AI Legal Index, its case rests on disclosure of the models underneath. It names GPT-5 and Claude Opus as the models and OpenAI and Anthropic as the providers, and states the zero data retention agreements negotiated with each. It names AWS, gives Canada and the United States as processing locations and publishes a full subprocessor list. It publishes no accuracy measurement, no rate and no indemnity running to the customer.
Questions buyers ask
Draftwise vs Spellbook: which is better for a transactional practice?
The difference between them is disclosure about the AI itself rather than about drafting. Spellbook names its models, providers and processing locations and publishes an independent EU AI Act opinion. Draftwise answers on the boundary instead, publishing private cloud and on premise deployment and reading the firm's own precedent through its document management system. From the AI Legal Index, based on each vendor's own published materials as of September 2, 2026. No vendor pays for placement.
Which one tells you which AI models it uses?
Spellbook. It names GPT-5 and Claude Opus as the models powering the product and identifies OpenAI and Anthropic as the providers. It describes the zero data retention agreements negotiated with each, names AWS as its primary cloud and states Canada and the United States as processing locations, with a full subprocessor list published. Draftwise names no model and no provider and publishes no subprocessor list. Its only reference to underlying models is a claim that customer data never trains public models. From the AI Legal Index, based on each vendor's own published materials as of September 2, 2026. No vendor pays for placement.
Can either run inside your own cloud?
Draftwise can. It publishes a private cloud dedicated to the firm in the customer's choice of AWS, Azure or GCP, with the firm's IT holding encryption keys and audit control, network access restrictable to the firm's own IP subnets, and no data leaving the VPC by default. Deployment into a firm's existing cloud or on premise is also offered. Spellbook publishes AWS hosting with data stored and processed in Canada and the United States, and publishes no customer selectable region or single tenant option. From the AI Legal Index, based on each vendor's own published materials as of September 2, 2026. No vendor pays for placement.
Do they work with iManage or NetDocuments?
Draftwise names both iManage and NetDocuments and describes what it does with them. Once the system is connected, it reads the firm's complete deal history including documents, clauses and tags and auto curates contract collections. Spellbook publishes a Word add in, Google Docs support and an integrations page, and no legal document management connector for iManage or NetDocuments. Neither publishes per integration documentation describing what moves in which direction. From the AI Legal Index, based on each vendor's own published materials as of September 2, 2026. No vendor pays for placement.
What do Draftwise and Spellbook both leave unpublished?
Neither publishes an accuracy figure, a hallucination rate or a test set for its own product. Neither publishes an indemnity running to the customer or a warranty on output. Neither names an ethics opinion, including ABA Formal Opinion 512. Neither documents what the product does when the precedent does not support the position requested. And neither publishes a rate at any tier, although Spellbook does publish its tier names and what each includes. From the AI Legal Index, based on each vendor's own published materials as of September 2, 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.
Draftwise's strongest published claim is that in a private cloud deployment no data leaves the firm's own VPC by default. No model or provider is named anywhere on the property, so that boundary cannot be reconciled against wherever the language model actually runs. Its own pages also differ on certifications. The home page states SOC 2 Type II and ISO 27001 certification, while the FAQ states Cyber Security Essentials Plus certification and SOC 2 compliance. On Spellbook, a published customer quote from a named partner states the product helps him bill an extra hour a day. Neither vendor publishes guidance on how AI assisted time should be recorded on a client bill. Neither vendor reviewed this page.