WWeave Bio

Weave Bio automates the paperwork of drug development. Its platform extracts data from study reports, drafts sections of investigational new drug (IND) applications, and drafts and tracks responses to questions from health authorities. In a study with Takeda, time to a first IND draft fell from about 100 hours to three. It is best for biotech and pharma regulatory teams, CROs and regulatory consultants facing tight submission timelines.

Weave Bio could help buyers by publishing pricing or a trial path; today both come through a sales process, and its accuracy figures, while strong, still leave work for expert reviewers. We rate it 7.3/10 on our scale. This review is based on Weave Bio’s press releases and its partners’ announcements; we have not used the platform ourselves.

Takeda and Parexel Put Numbers on Weave

Weave Bio is a small company, but two large partners, Takeda and Parexel, have put numbers on its results. Each figure below was checked on the source page.

For a buyer, the Takeda numbers are the most useful because they come from a named sponsor and a co-authored study rather than a marketing claim alone. A 97% cut in time to first draft is large, but read it alongside the 95% faithfulness figure: on a long IND section, one statement in twenty may not match the source, which is exactly what regulatory reviewers must catch. Weave’s pitch is that checking a traced draft is faster than writing from a blank page, and the study figures support that for first drafts.

The agency question figures are more modest. A 68% average answer quality score, inside a target of 50% to 70%, suggests the drafts are a starting point that saves search and assembly time rather than finished responses. That fits how Takeda’s chief regulatory officer describes the setup, with professionals retaining responsibility for content and final decisions.

One Platform for Regulatory Content From Data to Response

Weave sells one cloud platform for regulatory content preparation and lifecycle management. AI is applied at each step, from organizing data to authoring, review, publishing and responses to health authorities.

Preparing an IND With AutoIND

1. Load source data. Teams bring in preclinical study reports and other source documents.
2. Extract. The platform pulls the data each IND section needs from those documents.
3. Draft. It writes first drafts of sections, keeping traceability back to the source so reviewers can check each statement.
4. Review and publish. Regulatory experts verify, edit and approve, and the platform supports review and publishing of a submission-ready dossier.

Answering Health Authorities With HAQ Manager

After a submission, agencies send questions with tight deadlines. HAQ Manager extracts and tracks incoming questions, generates draft responses from source documents and past regulatory interactions, consolidates responses across functions for the final submission, and maintains version control so teams stay aligned. Weave built it with Takeda and launched it in November 2025, and the Takeda collaboration later extended to Response to Question workflows.

NDA, BLA and Ex-US Filings on the Roadmap

With its Series A, Weave said it would extend coverage to NDA and BLA submissions, health authority questions and post-market filings, expand beyond the FDA into Europe, Japan and Latin America, and add AI tools for portfolio insights.

The Partners Shaping the Product

Two partnerships explain much of how Weave has developed. Parexel, one of the larger CROs, became Weave’s design partner in September 2025 and contributes regulatory consulting across clinical, non-clinical, clinical pharmacology and chemistry, manufacturing and controls (CMC) work to improve the platform. Takeda co-developed HAQ Manager and co-authored the research paper on the IND phase. Working with a sponsor and a CRO gives Weave feedback from both sides of outsourced regulatory work, and it means many of the workflows were shaped by large-company processes, which smaller biotechs may need to adapt.

Expert Control Aligned With FDA Draft Guidance

Weave positions itself as assistance under expert control. Its approach follows the FDA’s January 2025 draft guidance on AI for regulatory decision-making and the January 2026 FDA and EMA guiding principles for good AI practice, both centered on human oversight. Regulatory professionals decide what is submitted, verify every claim against source and own communication with agencies.

Quoted Deals for Pharma, CROs and Consultants

Weave Bio does not publish prices. The platform is sold through its sales team to pharma companies, biotech firms, CROs and regulatory consultants.

Per-Program or Per-Submission, and HAQ Terms

Because every deal is quoted, the contract terms matter more than a price list. Ask whether pricing is per program, per submission or per seat, and whether HAQ Manager is included or a separate module. Confirm how unpublished study data is stored, whether it is used to train any model and which security certifications Weave holds. Ask which submission types and health authorities are supported today rather than on the roadmap, and whether a pilot on a single IND section is possible before a full commitment. Finally, check how documents can be exported if you leave, since regulatory content has to stay available for the life of a product.

Time Saved for a Small Biotech’s First IND

Without published prices, the useful comparison is time. These worked examples use the study figures, not a guarantee.

Small biotech filing its first IND. If five IND sections each took about 100 hours to draft by hand, that is 500 hours. At Takeda’s measured rate of about three hours each, first drafts would take 15 hours, freeing 485 hours, though expert review time comes on top.

Sponsor using a CRO. Parexel reported finishing INDs 50% faster with AutoIND. A sponsor outsourcing regulatory writing can ask its CRO whether Weave is in use and how the time saving shows up in fees.

Hidden and variable costs. Review effort does not disappear: with about 95% faithfulness to source, teams still need to check each statement. Implementation, onboarding and any work to bring in historical regulatory correspondence should be scoped in the quote.

Weave Bio Compared With Veeva AI Agents, IQVIA.ai and Medable AI Agents

Few listed agents target regulatory writing directly, so these are the closest life sciences platforms a sponsor is likely to weigh.

Weave is the specialist: it focuses on regulatory documents and agency questions, while Veeva and IQVIA add agents to broader platforms. Companies already running Vault RIM may prefer Veeva’s native agents; sponsors without a large regulatory stack, or CROs writing for many clients, are Weave’s natural buyers. More options are on our life sciences agents page.

Biotechs Writing an IND and Multi-Program Pharma

• Biotechs preparing an IND: AutoIND targets the preclinical IND, where small teams face the heaviest writing load.
• Pharma regulatory groups with many programs: Takeda’s 14-program rollout shows the platform working at portfolio scale.
• CROs and regulatory consultants: Weave sells directly to them, and Parexel uses it as design partner.
• Teams buried in agency questions: HAQ Manager tracks questions and drafts responses across functions.
• Organizations that need auditable AI use: source traceability and human sign-off align with FDA and EMA principles; see also our healthcare agents.

Discovery Scientists and Clinical Ops Teams

• Discovery scientists: research agents such as Biomni Lab or Kosmos fit lab analysis better.
• Clinical operations teams: trial build and monitoring are better served by Medable.
• Companies standardized on Veeva Vault: native Veeva AI Agents avoid another platform.
• Buyers who need a trial before talking to sales: Weave Bio does not yet offer a self-serve option.
• Teams filing mainly outside the US today: Europe, Japan and Latin America coverage is still being built out.
• Teams that want a fully automated writer: Weave is designed for drafting and tracking under expert control, so it will not remove the need for experienced regulatory writers and reviewers on each submission.

95% Faithfulness Still Means Expert Checking

• Accuracy figures: about 95% faithfulness to source and a 68% answer quality score mean every draft needs expert checking.
• Security documentation: Weave’s announcements do not yet list certifications; request its security and data handling terms before uploading unpublished study data.
• Partner-reported results: the Takeda and Parexel figures come from specific use cases and may not match your programs.
• CRO exclusivity: Parexel is the only CRO licensed to use new Weave products for a period after launch.
• Scope still growing: NDA, BLA and post-market coverage were listed as expansion goals in late 2025.
• Young vendor: founded in 2022 with $36 million raised, so plan for contract terms that protect your data if the company changes hands.

Takeda Expands to 14 Programs

On September 1, 2026, Weave Bio announced an expanded collaboration with Takeda covering 14 active programs, after a co-authored study found a 97% reduction in time to first IND draft and more than 100 agency questions handled in its HAQ and RTQ deployment.

On November 6, 2025, Weave launched HAQ Manager for health authority questions, developed with Takeda. On October 16, 2025, it raised a $20 million Series A led by USVP. On September 30, 2025, Parexel announced an AI partnership with Weave, naming it Parexel’s CRO design partner.

Well Evidenced for IND Teams, With Security Detail to Add

For regulatory teams and CROs preparing INDs or answering agency questions under deadline, Weave Bio is one of the better-evidenced AI options. It scores 7.3/10 on our scale. The Takeda study, the 14-program rollout and Parexel’s 50% faster INDs are stronger proof than most life sciences AI vendors offer, and the human-in-control design matches where regulators are heading.

The score reflects what a buyer cannot yet see in advance: prices, a trial, named security certifications and public ratings, each of which Weave Bio could add. Ask for a pilot on one IND section you have already written, compare drafting time and the number of corrections needed, and get security terms in writing. For a small biotech, the deciding question is whether the hours saved on a first IND outweigh a license fee you cannot see in advance; for a CRO or a sponsor with a dozen programs, the Takeda rollout suggests the math is more likely to work. Either way, measure the correction rate on your own documents, since that is where the 95% faithfulness figure turns into real reviewer hours. Compare more options on our life sciences agents page.

Weave Bio for Regulatory Writers, Plain Answers

How Much Does Weave Bio Cost?

Weave Bio does not publish prices. The platform is quoted by its sales team for pharma companies, biotechs, CROs and regulatory consultants.

What Is AutoIND?

AutoIND is Weave’s flagship module for preclinical IND preparation. It extracts data from study reports and drafts IND sections with traceability to source. Parexel reported completing INDs 50% faster with it.

Does Weave Bio Submit Documents to the FDA on Its Own?

No. It drafts, organizes and tracks content, and helps publish submission-ready dossiers, but regulatory professionals review, approve and submit. Weave describes its approach as following FDA and EMA principles on human oversight of AI.

How Accurate Are Weave Bio Drafts?

In the Takeda study, faithfulness to source data was about 95%, and agency question responses averaged a 68% quality score against a 50% to 70% goal. Drafts save time but need expert review.

Which Companies Use Weave Bio?

Takeda is deploying it across 14 programs and co-developed HAQ Manager, and Parexel uses it as its CRO design partner. Weave says it serves biotech and pharma companies, CROs and regulatory consultants.