Fearn Launches AI-Native Patent Firm to Fix Legal Billing

The News

Fearn, an AI-native patent prosecution firm founded by former Morrison & Foerster attorney Han Kim and Caltech AI PhD Angela Gao, has launched publicly with $5.5 million in seed funding from Kindred Ventures, a16z Speedrun, Designer Fund, and Essence VC. The firm pairs former Big Law patent experts with a proprietary drafting and portfolio management system called FearnOS, which represents patent applications as structured graphs rather than linear documents, enabling provisional filings in as little as three business days at a flat fee of $2,500. Fearn targets the $14 billion global patent market, specifically the roughly 150,000 annual patent applications filed by early-stage companies, with non-provisional applications priced at $9,000 including USPTO fees and a money-back guarantee if no claims are allowed.

Analyst Take

The economics are the product

Fearn’s core thesis is not that AI can draft patents faster. It’s that traditional law firm economics are structurally incompatible with AI efficiency gains. When an attorney bills by the hour, faster work is a revenue problem, not a selling point. Fearn resolves this tension by building a fixed-fee firm around proprietary tooling from day one, so speed and margin move together rather than against each other. The reported gross margins above 80% on a $9,000 non-provisional filing, compared to market rates of $18,000–$40,000, suggest this model can sustain itself without the utilization math that governs traditional practices. That’s the genuine structural innovation here, not the AI itself.

FearnOS adds a technically meaningful layer on top of that business model shift. By representing a patent as a graph mapping claims to supporting disclosure, figures, and source material, the system preserves the relational structure that makes patent prosecution hard to automate well. Claims don’t exist in isolation; they depend on what’s disclosed, and a change to one element can affect validity across the application. Building a data model that tracks those dependencies, while maintaining a full audit trail of attorney edits, is a more defensible technical choice than a simple document-generation wrapper. Developers building legal tech or document-intelligence systems will recognize this as a deliberate graph-based knowledge representation decision, not a generative AI demo dressed up as software.

Where the government sector data points to a parallel tension

The structural problem Fearn is solving in legal services has a direct analogue in public sector technology procurement. ECI Research’s Google GovTech Survey found that 56.0% of respondents said procurement or contractual requirements “frequently” force their engineering teams to use suboptimal developer tools, specifically because approved vendor lists lack modern developer platforms. The mechanism is nearly identical: incumbents with established contract vehicles hold market position not because their products are superior, but because the procurement structure rewards longevity over performance. Fearn is betting that startups, at least, will route around that same dynamic in legal services by choosing on speed and price transparency rather than firm reputation alone.

The AI governance dimension also deserves attention. Fearn’s decision to build privately hosted infrastructure for clients like Photon Spear, a defense technology company, aims to address the data sovereignty concern that slows AI adoption in regulated industries broadly. According to ECI Research’s Google GovTech Survey, 31.8% of respondents identified “FedRAMP/compliance approval friction for AI vendors” as the single largest blocker preventing widespread AI adoption in developer workflows. Patent prosecution involves pre-filing IP that is, by definition, the most sensitive material a startup possesses. Fearn’s architecture, where sensitive technical documentation connects to the platform through controlled, private integrations rather than multi-tenant cloud endpoints, is a response to that concern, even if the customer base is commercial rather than government.

The quality guarantee changes the risk calculus

One detail worth isolating for ITDMs evaluating any AI-assisted professional services vendor: the money-back guarantee on drafting fees if a non-provisional receives no allowed claims. This is outcome-based pricing applied to legal work. It shifts a portion of execution risk back to the vendor, which is structurally unusual in professional services. ECI Research’s Google GovTech Survey found that only 15.6% of respondents said value-based or outcome-based contracting is the most common vehicle used for their application development initiatives, despite widespread acknowledgment that time-and-materials contracts (used by 56.2%) misalign incentives on software delivery. Fearn’s guarantee doesn’t make it a government contractor, but it does model a pricing structure that the broader technology procurement world is moving toward, slowly.

Looking Ahead

Fearn’s near-term expansion into patent families, international filings, office actions, and portfolio strategy is where the business model will be tested at scale. Drafting a first provisional quickly is a tractable AI problem. Managing a growing portfolio across jurisdictions, with evolving claim scopes, prior art landscapes, and prosecution histories, is substantially harder. If FearnOS can maintain its graph-based coherence across that complexity, the firm becomes a platform business with compounding data advantages. If it can’t, the firm risks becoming a low-cost drafting service that struggles to retain clients as their IP needs mature.

The broader competitive response is already forming. Large firms are investing in proprietary AI infrastructure precisely because they recognize that off-the-shelf tools commoditize their junior associate leverage. But their cost structures, partnership distributions, real estate, and recruiting pipelines built around the billable hour, constrain how far they can move on pricing even if they achieve comparable technical results. Fearn’s $5.5 million is a small number relative to the incumbents it’s challenging, but the structural advantage is not capital. It’s that Fearn has no legacy model to protect. That asymmetry has defined every major professional services disruption, and there’s no reason patent law should be an exception.

Authors

  • Paul Nashawaty

    Paul Nashawaty, Practice Leader and Lead Principal Analyst, specializes in application modernization across build, release and operations. With a wealth of expertise in digital transformation initiatives spanning front-end and back-end systems, he also possesses comprehensive knowledge of the underlying infrastructure ecosystem crucial for supporting modernization endeavors. With over 25 years of experience, Paul has a proven track record in implementing effective go-to-market strategies, including the identification of new market channels, the growth and cultivation of partner ecosystems, and the successful execution of strategic plans resulting in positive business outcomes for his clients.

    View all posts
  • With over 15 years of hands-on experience in operations roles across legal, financial, and technology sectors, Sam Weston brings deep expertise in the systems that power modern enterprises such as ERP, CRM, HCM, CX, and beyond. Her career has spanned the full spectrum of enterprise applications, from optimizing business processes and managing platforms to leading digital transformation initiatives.

    Sam has transitioned her expertise into the analyst arena, focusing on enterprise applications and the evolving role they play in business productivity and transformation. She provides independent insights that bridge technology capabilities with business outcomes, helping organizations and vendors alike navigate a changing enterprise software landscape.

    View all posts