Hebbia is an enterprise-first AI platform focused on knowledge work across finance, law, and corporate functions. Its core product, Matrix, is presented as an AI interface/architecture for multi-step workflows, large-context reasoning, and multi-modal inputs. Primary value propositions emphasized on the site include supporting complex multi-step tasks, handling very large/multi-modal data (large or effectively infinite context), providing traceability/citations for verifiable AI actions, and enterprise security and compliance. Publicly cited scale and customer metrics on the site include claims of 1,000+ live use cases, 2T+ tokens processed, $21T AUM across client firms, and 5+ years deployed at top firms. Key capabilities and examples of use cases presented publicly: finance workflows (market ramp-on, earnings-call analysis, memo drafting/review, credit agreement term extraction, deal benchmarking, risk screening, diligence automation), legal and corporate workflows (contract and lease analysis, extracting deal points, automating first drafts and summaries, regulatory monitoring), multi-modal reasoning (text plus charts/graphs), large-scale document processing, editable prompts/agents for continuous improvement, and citations for verifiability. The site positions Hebbia as replacing narrow chat workflows with auditable, agent-style workflows that scale across organizations and aim to deliver measurable value quickly. Security and trust information referenced publicly: SOC 2 Type I and SOC 2 Type II are referenced (Trust Center shows SOC 2 Type II LoA with a June 2025 reference), GDPR and CCPA readiness are mentioned, and encryption in transit and at rest is cited (AES-256 at rest and TLS 1.3 in transit are referenced on the site). Hebbia states it does not train on customer data and asserts data ownership remains with customers. The Trust Center (trust.hebbia.ai) and security pages are referenced as the place to obtain detailed artifacts and attestations (SOC reports, penetration test summaries, DPA, etc.). Pricing and purchasing model observed: No public, line-item pricing or free-trial plans were found. The pricing page emphasizes enterprise customization and ROI, and the site repeatedly directs potential customers to "Book a demo." The public evidence indicates pricing is negotiated/custom enterprise licensing rather than a public self-serve plan. Integrations and technical connectivity: The product pages and blog reference a "wealth of integrations" and connectors to data sources, but no single public API documentation or exhaustive connector list was discovered on public marketing pages. The available evidence suggests connectors or API access may be provided to customers or discussed during demos, rather than published exhaustively on the public site. News and company updates found: a blog post confirming a $130M Series B led by a16z (July 2024) and product announcements such as "Introducing Matrix" plus posts referencing GPT-5/Matrix and other product development articles. Public links reviewed include the home/overview, product/Matrix pages, pricing, demo booking, security page, trust center, industry (finance) page, several blog posts, and careers pages. Public gaps identified: no line-by-line public pricing or free-trial details, no complete public list of integrations/connectors or technical API/docs on marketing pages, and no public SLA, detailed data retention policy, or explicit deployment options beyond general security claims. Detailed SOC 2 reports or penetration test results appear to be accessible via the Trust Center or available on request rather than embedded on marketing pages. Recommended next steps publicly suggested by the reviewer: book a demo with a live walkthrough using your data or a POC use case and request supported connectors/APIs and implementation timelines; request security artifacts from the Trust Center (current SOC 2 Type II reports and attestation letters, penetration test summaries, DPA, data retention/deletion/export procedures); ask for commercial details (pricing model, licensing metrics, SLA terms); clarify technical questions (deployment options, data residency, encryption key management, sandbox/test environments, custom agent deployment/management); and request references or case studies aligned to your use case and scale needs.