By Chris Skoff, Managing Director, Marks Baughan
The intellectual property (IP) software industry has seen a torrent of early and growth stage investment activity in the year-to-date 2025. Notable and intriguing players include Solve Intelligence, DeepIP, Vermillio, Patlytics, &AI, Podqi, and Cypris AI — all of which are raising millions for their AI-driven technologies.
This influx of activity demonstrates the increasing demand — and the many needs and use cases — for AI throughout IP, within innovation management; patents and trademarks; branding and brand protection; trade secrets; music, media and publishing copyrights; and other intangibles. As we watch investors place and increase these early bets, let’s take a closer look at how AI is transforming IP so both innovators and investors can understand the implications of this technology that is evolving at a break-neck pace.
Marks Baughan tracks market activity and players throughout the IP software industry, and we see three areas within IP surfacing among AI-based startups:
- AI-driven portfolio analytics and competitive intelligence
- AI-driven workflows — from innovation and ideation through IP formation, prosecution, and maintenance
- AI-driven patent application drafting and office action responses
The first theme has ushered in tools that give companies a competitive edge, and the next two offer new ways to discover efficiencies, reduce error rates, and cut costs associated with managing large patent and trademark portfolios. These areas are heating up because they represent a highly disruptive shift to business as usual.
The CEO’s perspective on the AI-in-IP moment
From Patlytics CEO Paul Lee’s point of view, this disruption is about a movement of labor “from law firms and consulting companies back to the corporate in-house.” Speaking to Marks Baughan, Lee revealed, “Such a labor movement has never happened before at this magnitude. Additionally, law firms and consulting companies want to be more efficient themselves to see greater ROI from their billing rates and hours. That’s why the market is seeing such increasing momentum in this trend right now.”
Steve Hafif, CEO of Cypris AI, offered additional insight about IP areas ripe for AI, namely “markets with a lot of data that can be easily used in GPT/RAG-based systems, which then fit into common workflows that already exist.” A prior art search is an example of this type of use case where publicly available data integrates with a large language model (LLM) to support a chatbot experience that eliminates the need for manual search.
Hafif said, “There are many examples of use cases like that in the IP space where manual research processes done in house or by lawyers and external firms can be completely automated. Any amount of money you’re spending there can be drastically reduced with AI.”
In R&D, Hafif is seeing AI implemented into technical strategy. “A vertical system that’s trained on a very specific dataset allows you to describe the type of concept or technology you want to build, and the system will reference prior art to offer a technical strategy on how to build that type of innovation,” he said.
In trademarks, Haloo’s CEO Julie MacDonell says the primary application of AI that will prove “massively disruptive” is the automation of trademark prosecution work. “Trademark clearance, watch, and workflow have long been high-volume, high-friction tasks — still dependent on humans for accuracy,” she told Marks Baughan. “AI is changing that, unlocking real efficiency in managing the legal and procedural steps of securing and maintaining trademarks at scale.”
Haloo, which offers SaaS for IP teams going AI-first and has developed APIs for self-serve trademarking, has caught the attention of investors for this reason and more.
“There are 400 million small businesses in the world. Most will never file a trademark — not because they don’t want to, but because they’ve never had access,” MacDonell said. “When we remove that barrier, we enable a market that didn’t exist before. This is a huge technology-enabled market transformation.”
As AI transforms nearly every aspect of legal tech, these are the three major IP domains to watch:
AI-driven portfolio analytics and competitive intelligence
This area currently has the most investment activity of the three. Numerous start-ups are developing AI tools that spot holes, opportunities, and/or identify competitive threats in patent portfolios. Organizations use this data to make investment decisions that remedy gaps and weaknesses in their portfolio or to discover opportunities to monetize their existing IP more aggressively. AI overhauls the traditional patent and trademark search methods — adding speed, semantic relevance, risk prediction, global coverage, and more to results produced by human researchers.
Patlytics’ closing of a $14 million Series A round2 in February exemplifies the fervor for these solutions. The company offers an AI-powered patent workflow platform that performs a number of functionalities — complex portfolio comparisons among them. The tool can analyze an entire patent portfolio to deliver the intelligence decision-makers need to better understand their competitors and market. The platform also provides predictive analytics for patent validity and infringement to help companies improve IP risk management and strategize research and development (R&D) investments.
Providing a tool that applies to a number of use cases and solves multiple problems out of the gate was intentional, Lee says, and it’s a big reason Patlytics is seeing such investor activity right now. He brought what he calls a compound startup to market to deliver on a major transformational capability of AI: the ability to bundle separate use cases together to improve multiple operational components with one tool.
Patlytics is unique in that it didn’t go to market with one product and subsequently add layers of functionality for various use cases. “When you add different layers on, you’re dealing with data architecture revisions, and the setup is inefficient from the beginning,” said Lee. His company’s solution is inherently modular, enabling the seamless application of its capabilities to dozens of needs across IP. The product does so without putting work back on the individual — Patlytics’ core focus.
“People don’t want more work thrown at them under the guise of taking work away, like with dashboarding tools that require a bunch of analytics,” said Lee. “Fundamental to Patlytics is the concept of taking work away from users to drive real outcomes,” he added.
AI-driven workflows
Software that automates administrative and paralegal workflows — and augments those staffs’ capabilities — is garnering investor attention. These workflows include:
- Auto-docketing and docketing system maintenance
- Managing files
- Performing routine correspondence
- Coordinating reminders, tasks, and alerts with attorneys, clients, agents, and foreign patent offices
Tools that automatically record, track, and manage deadlines, events, and tasks allow organizations to redeploy paralegal and IP department resources for higher-value work — or cut some resources altogether. By processing docketing items within seconds of receipt from patent offices, organizations improve efficiency and accuracy in managing IP portfolios.
Black Hills AI is a leader in this domain with its SureDocket™ service that claims to automate up to 90% of U.S. patent docketing tasks. The solution processes incoming correspondence in real time, ensuring docket entries are accurate and timely. The company also offers DocketSaver™ — an automated verification tool that cross-references U.S. Patent and Trademark Office (USPTO) correspondence with docket entries to ensure actions are recorded correctly and identify any mistakes before damage is done.
Interestingly, Black Hills also announced3 a partnership with DeepIP in October 2024 that allows Black Hills AI users to “significantly enhance their IP management processes” with DeepIP’s AI capabilities.
Haloo has docketing on the brain as well. MacDonell mentioned her company’s long-term vision includes docketing automation aimed at relieving trademark practice of so much human-based paralegal work. “Docketing will become highly automated,” she said, “because it’s structured, repetitive, and rule-based.”
She mused about a future that includes “a deep kind of agentic AI” where the technology “performs whole workflows in docketing — and does so very accurately.”
“It’s one of those things that is going to be difficult to get right,” she added. “But the companies that are really going to put the R&D into it, hire very strong teams, and really focus on data quality will carry docketing into the future.”
AI-driven patent application drafting and office action responses
We’re seeing heavy investment in tools that enable users to automate patent drafting and office action responses and management. Startups aim to overhaul the notoriously time-consuming, expensive, and tedious application filing process with AI-based solutions that slash the effort required. These tools offer AI-driven document creation that automates office action responses, invention disclosure analyses, and new patent applications. Such automation allows companies and law firms to redirect attorney and paralegal hours to more profitable activity, achieve more efficient and supportive paralegal work, and dramatically cut spending on patent drafting.
To an extent, these tools also have the capability to optimize claims language tailored to specific technologies and even patent examiners — including words certain examiners may favor and avoiding ones they may disfavor, for example. For new paralegals and patent attorneys who are getting up to speed and learning these nuances, this capability is a huge boon. It also helps senior attorneys and partners rest easy knowing the technology acts as an extra set of (machine) eyes.
In this arena, Solve Intelligence — provider of an AI-driven, in-browser document editor that helps write patents — raised $12 million4 in a Series A funding round in early April. A few weeks prior, DeepIP secured $15 million in Series A funding5 for its generative AI-based patent drafting tool that purportedly saves IP attorneys 50% of the time they’d spend manually drafting applications.
DeepIP also unveiled a new product module in February designed to “transform the traditional, labor-intensive process of responding to USPTO office actions” according to the company.6 The tool enables users to more efficiently and effectively manage office actions, prioritize and speed up responses, and refine arguments.
Hafif sees a specific future for this realm of work as “phase two” of AI implementation rolls in. Where phase one is using AI to draft documents and perform research, phase two is letting “AI be more agentic in that it actually does the tasks you’re trying to do,” he said.
“Are you ready to let an AI agent communicate with the USPTO and then draft and submit the patent? That’s where it’s going, and many people don’t know phase two even exists,” he added.
All eyes on AI in IP
In truth, much of the legal world is still eyeing phase one with concern. MacDonell noted that lawyers’ resistance to change and sluggish tech adoption aren’t serving them in this new era of AI. In the trademark space specifically, AI is a boon for lawyers: “When you democratize access [to trademark work], you give people a better chance at building a company that can actually get off the ground,” she said. “You’re giving them legal footing to do that, which then grows the need for lawyers.”
As Haloo pursues its holy grail — to become the infrastructure layer for trademarks globally — MacDonell hopes lawyers will see the opportunity afforded by AI. “The technology will grow the market and bring huge returns for lawyers,” she said. “The ones that get in early will be the ones that thrive.”
For Lee, the holy grail of AI is “reducing transactional efficiency to a flat line, whether it’s in litigation, prosecution, docketing, or internal. If you can pull that off, that’s when things get extremely interesting.”
Hafif’s holy grail is when “AI understands the tribal knowledge of the organization, so every time you log into your GPT product, you’re not starting from zero. You’re starting from two, and then four in your next session, and so on. It keeps a memory of the tribal knowledge of thousands of people in the organization, which it builds on to answer queries as the market evolves. You want to have a question in September that the AI realizes it can answer in February as new research and studies come out.”
As AI-driven startups innovate to reach these ends, they draw the intense attention of large incumbents in the IP space — Anaqua, Questel, CPA, and Corsearch to name a few — all of whom are also investing in their own AI initiatives. Indeed, Anaqua, with its new capital partner Nordic Capital, isn’t wasting time going after AI-powered solutions. In May 2025, the company announced7 its acquisition of RightHub — a provider of an “AI workbench” that allows patent attorneys and in-house IP managers to “use AI to focus on more strategic aspects of patent prosecution.”8 For example, RightHub’s AI-powered auto-docketing solution enables users to auto-generate and update case files from documents and emails, eliminate inbox-jumping, reduce human error, and free up time for higher-value IP tasks.
Anaqua has been 100% patent-focused for 15 years, so this acquisition bolsters its bread-and-butter, but will AI push incumbents to focus on areas outside their traditional core offerings? Anaqua CEO Justin Crotty recently brought up some intriguing mentions of the company’s work in AI, innovation, and branding during a conversation with VentureFizz.9 Perhaps the interview signals Anaqua’s opening of other growth vectors.
As the AI juggernaut forges ahead, winners and leaders will continue to emerge from the start-up community — forcing large incumbents to engage in an ongoing “buy versus build” calculus. Marks Baughan lives in the IP space and keeps a keen eye on emerging AI leaders. We track who is developing what, who is gaining traction, and where the investors and acquirors are focusing strategic attention — and we deliver that market intelligence to our clients. Learn more about our work here.
Sources:
- PitchBook
- Patlytics secures $14 million Series A led by Next47 • Patlytics. (n.d.). https://www.patlytics.ai/blog/patlytics-secures-14-million-series-a-led-by-next47
- Business Wire. (2024, October 22). Black Hills AI and DeepIP announce Strategic Integration Partnership. Yahoo Finance. https://finance.yahoo.com/news/black-hills-ai-deepip-announce-130100191.html
- Solve Intelligence raises $12M Series A to build AI for patents | Solve Intelligence. (n.d.). https://www.solveintelligence.com/blog/post/series-a-funding-to-build-ai-for-patents
- DeepIP raises $15M to revolutionize AI patent drafting. (n.d.). https://www.deepip.ai/blog/deepip-raises-15m-ai-patent-drafting
- Introducing DeepIP’s AI Office Action Module for enhanced patent practice. (n.d.). https://www.deepip.ai/blog/ai-office-action
- Anaqua IP Management Software and Services. (2025, May 15). Anaqua acquires RightHub to accelerate global growth – Anaqua IP Management Software and Services. https://www.anaqua.com/resource/anaqua-acquires-righthub-to-accelerate-global-growth/
- RightHub. (n.d.). https://righthub.com/ai
- VentureFizz. (2025, April 9). Discover Anaqua: A leading provider of IP management software and services [Video]. YouTube. https://www.youtube.com/watch?v=T1YimvLDRik