Hudson Labs: Coverage Update — 21 Oct 2026

GAII Intelligence defines the AI Research & Intelligence market as software that applies generative and agentic AI to the research process of professional investors: locating candidate securities, assembling and interrogating evidence, producing written analysis with traceable sources, and monitoring positions once held. Buyers are hedge funds, long-only managers, multi-manager platforms, family offices and allocators; the economic buyer is typically the chief technology officer, the head of research, or a portfolio manager with a technology budget.

Capabilities we treat as mandatory

• Evidence retrieval at scale: the ability to search and reason across regulatory filings, transcripts and news, and across documents the client supplies.
• Source traceability: every material claim in an output must be reachable back to the passage that supports it.
• Institutional data handling: a stated position on where client data goes, how long it is held, and whether it is used for model training.

Capabilities we treat as differentiating

• Persistent state: the system remembers a thesis, a coverage list or a position between sessions and acts on it without being re-prompted.
• Workflow surfaces: delivery inside the applications analysts already use — the spreadsheet, the research management system, email and messaging.
• Owned or exclusive content: material that a competitor cannot obtain by signing the same third-party contract.
• Published evaluation: a measured, disclosed accuracy methodology rather than an assertion of quality.

Hudson Labs competes in the filings and forensic-analytics corner of the research segment. AlphaSense and Hebbia are the broad platforms buyers will compare it with; for forensic and guidance work its closest alternatives sit outside our rated cohort. It is most useful as a short-side and risk overlay alongside a generalist research tool.

The following are the capabilities and characteristics that, in our assessment, most distinguish Hudson Labs within the AI Research & Intelligence market. Each is drawn from publicly documented product capability, published interviews with management, or independent hands-on review.

• Forensic depth: Fraud-likelihood scores across accounting, related parties, internal controls and governance, built by a CPA founder and refined since 2021; its findings have repeatedly been cited by financial press.
• Commercial transparency: Published Core pricing, a 14-day trial, an indicative Institutional price and per-project Deep Dive pricing. No other company in our coverage is as open about cost.
• Accountability on numbers: A $50 refund for each confirmed numeric hallucination, restatement-adjusted five-year history and verbatim cited excerpts.
• Contractual no-training clause: The institutional terms prohibit using customer content to train, fine-tune or improve any model, with published retention periods and ten named subprocessors.
• Shipping cadence: At least nine dated launches between March and August 2026, including Market Intelligence, Agents, Deep Dive, Teams delivery and MCP in Claude.
• Applied-AI leadership: The CTO wrote O'Reilly's Designing Large Language Model Applications and worked on BigScience BLOOM.
• Independent accuracy signal: A third-party review rated multi-period extraction accuracy high.

The following are factors that prospective clients should weigh, and areas where our public review did not find sufficient evidence to reach a settled view. Several are common across vendors at this stage of the market and are not specific to Hudson Labs. We encourage prospective clients to raise each of them directly with the company; we have done the same, and will update this note as answers arrive.

• Scale and capital: Company-filled sources show roughly five to eleven staff, and the only financing located is about $1.2m from 2021. Buyers should weigh continuity and key-person risk.
• Security attestation: Hudson Labs describes its controls as aligned with SOC 2; we did not locate an attestation, ISO 27001, a trust centre, SSO/SAML/SCIM or an SLA.
• Content scope: Coverage centres on US EDGAR filings, transcripts and decks with S&P consensus. Expert calls, broker research, newswires and non-US filings are not offered, as the company itself notes.
• Internal data: Document upload was unsupported in January 2026 and connectors were in beta or coming soon in June; buyers should confirm general availability.
• Tier gating: Forensic risk, export, sharing and MCP sit in the quote-only Institutional tier, so self-serve pricing does not reflect a full deployment.
• Model provider: Query content is sent to OpenAI; a zero-data-retention arrangement and MNPI handling were not located.
• Inconsistent figures: Client AUM, Core price and penetration-test cadence differ across the company's own pages.
• Evaluation evidence: Accuracy claims are vendor-run; no third-party benchmark score is published, and the refund guarantee excludes calculations and interpretation.

Hudson Labs Inc., formerly Bedrock AI, was founded in Toronto in 2019 by Kris Bennatti, a chartered professional accountant and former KPMG senior accountant and data scientist, and Suhas Pai, formerly a staff software engineer at IBM. It launched commercially in April 2021, joined Y Combinator's summer 2021 batch and rebranded as Hudson Labs in December 2023.

Hudson Labs raised about $1m in a 2021 seed round from investors including Good News Ventures and Born Capital, after Y Combinator's initial investment; no later round has been located. Its institutional terms are governed by New York law. The company reports serving hundreds of institutional investors, and names Princeton Value Partners, Telos Research and Fundamental Edge among users.

Company-reported figures are shown as such and are not independently verified. Verification is carried into the vendor questionnaire.

Hudson Labs presents a research platform centred on SEC filings, with forensic scoring and monitoring agents. Figure 2 sets out our assessment of how it maps onto the investment workflow.

Co-Analyst

Natural-language research across filings, transcripts, decks and press releases with verbatim citations.

Guidance and earnings

Capture of hard, soft and implied guidance, and earnings summaries against consensus.

Forensic Risk and Market Intelligence

Fraud-likelihood scores and market-wide screens on fundamentals, sentiment, insider sales and comment letters.

Agents and Deep Dive

Topic Agents on any SEC filing event delivered by email or Teams, and long-running parallel research projects priced per project.

Scope of the product today

We did not locate an Excel add-in or a public REST API. MCP is available in Claude for Institutional plans only.

Hudson Labs combines purpose-built models trained on financial documents with OpenAI models, using context engineering, topic spans and proprietary relevance ranking, and returns verbatim cited excerpts. Guardrails use Guardrails AI and NVIDIA NeMo.

Evaluation is the area to probe. The company cites third-party benchmarks of frontier models but publishes no score of its own, and its head-to-head comparisons are vendor-run.

Coverage spans all US EDGAR filers and ADRs — 10-K, 10-Q, 8-K, proxies, S-1 and 20-F filings, transcripts, decks, ownership filings and comment letters — with restatement-adjusted five-year history and consensus from S&P Market Intelligence.

Expert calls, broker research, newswires and non-US filings are not offered. Client document upload and connectors were moving from beta in mid-2026.

Hudson Labs fits the long-short equity analyst, particularly on the short side: forensic screens, guidance tracking and filing-event agents delivered by email or Teams.

It is not a modelling or deliverable tool. We did not locate an Excel add-in, and output leaves the platform by export.

Hudson Labs publishes a contractual no-training clause, retention periods, ten named subprocessors, AWS and Azure hosting in the US and Canada, TLS 1.2 and AES-256 encryption, and regular penetration testing with results under NDA. Figure 3 sets out what our public review located.

We would encourage any firm evaluating Hudson Labs to ask about SOC 2 attestation plans, SSO/SAML, an SLA, zero-data-retention terms with OpenAI and MNPI handling.

The offsetting factors are real. The no-training clause is contractual rather than marketing, and the retention and subprocessor disclosure is more specific than most peers publish.

Hudson Labs publishes its Core price — about $99 a month billed annually — with a 14-day trial, and indicates Institutional pricing from about $15,000 per team per year. Deep Dive projects cost $25 to $50 or more each.

Prices differ slightly across the company's pages, and key features sit in the Institutional tier. Buyers should confirm the cost of the configuration they need.

Hudson Labs sits in the Emerging quadrant of the GAII Intelligence Matrix, held below both lines by data breadth, enterprise evidence and scale.

GAII coverage universe, 21 October 2026. Scores as currently published on the GAII Intelligence page; preliminary and based on public information.

11.1 Where Hudson Labs is strongest relative to peers

• Against broad research platforms, on forensic accounting depth and guidance extraction.
• Against every peer, on price transparency and a trial anyone can start.
• Against general assistants, on restatement-adjusted numbers with a refund guarantee.

11.2 Areas of narrower functional focus

Hudson Labs has built depth in US filings and forensics rather than breadth of content or enterprise scale.

• Content: no expert calls, broker research or non-US filings.
• Enterprise: no SOC 2 attestation, SSO or SLA located.
• Delivery: no Excel add-in or public API located.
• Scale: very small team and seed-stage capital.

Scores are on a 0–10 scale. Each dimension score is the mean of its sub-criteria, and the GAII Overall Score is the weighted mean of the six dimensions. The detailed sub-criteria scores that build each dimension are set out in Appendix A.

These are the areas where our public review did not find sufficient evidence to form a settled view. We have put them to the company and will update this note when they are answered. Prospective clients may find them a useful agenda for their own evaluation call.

• Whether a SOC 2 Type I or Type II attestation exists or is in progress.
• Availability of SSO/SAML, SCIM and an SLA.
• Zero-data-retention terms with OpenAI.
• MNPI handling policies.
• Current headcount and engineering team.
• Whether capital has been raised since 2021.
• General availability of uploads and connectors.
• Plans for an Excel add-in, API or MCP beyond Claude.
• Forensic Risk Score methodology and backtest.
• Current Core and Institutional pricing.
• Number of paying institutional clients.
• Penetration-test cadence and latest date.

This note is an assessment of a vendor against a defined framework, not a recommendation to buy or to decline. The following are the considerations we would put at the centre of an evaluation of Hudson Labs specifically.

If you are a fundamental equity manager

• Use the trial to test forensic screens on names you know well.
• Pair with a generalist platform for expert and broker content.
• Check guidance capture against your own notes for a quarter.

If you are responsible for technology or vendor risk

• Open with Figure 3; the no-training clause is a strong starting point.
• Ask about SOC 2 plans before enterprise use.
• Confirm OpenAI data handling.

If you are assessing the category rather than the company

• Price transparency remains rare; Hudson Labs shows it is possible.
• Accountability mechanisms such as refunds are a new way to evidence accuracy.
• Specialist overlays sit alongside, not instead of, generalist research platforms.

Each dimension score in Section 12 is the mean of the sub-criteria below.