Invention Harvesting: Leveraging AI for Greater Impact

“Why not go out on a limb? Isn’t that where the fruit is?”
~ Frank Scully, American journalist

Back in 2016, I wrote an article for IPWatchdog describing invention harvesting, impediments in its way and best practices for overcoming them. Based on colleagues’ feedback, the article remains relevant for intellectual property professionals and innovators, including those starting their harvesting journey and others seeking to optimize strategies.

Recent advances in artificial intelligence (AI) and related software tools can be leveraged by companies, universities, law firms and IP service providers to overcome common impediments to invention harvesting.

Adopting and embracing such tools is a powerful additional best practice for achieving success in invention harvesting.

Invention Harvesting Fundamentals

Invention harvesting is a framework for (a) prompting innovators to conceive of new inventions or describe inventions already conceived, but not yet appreciated or articulated, and (b) documenting such inventions for subsequent protection via the patenting process or, if appropriate, as trade secrets.

Other descriptors for invention harvesting include invention capture, mining, and brainstorming; ideation, innovation, discovery, or patent capture; innovation gathering; idea collection; and patent portfolio building.

Harvesting may reap innovations arising from active R&D projects, as well as innovations not tied to projects. Such non-project innovations may include inventions made by inventor-employees on an ad hoc basis or as part of group innovation challenges, hackathons, sprints, or jams. Ideas or solutions may be elicited to address technical, cost, or user challenges and frustrations, or to push the envelope on futuristic, blue-sky thinking.

Invention harvesting can be formal or informal in approach and may be directed to one or multiple inventions and/or topical areas. It’s often led or fostered by in-house IP counsel and/or outside IP counsel, such as when IP asset generation is a primary goal. It also can be inventor, engineering leader or marketing leader-driven, where teams care most about brainstorming technologies for incorporation in existing or future products and services; new IP may be considered icing on the cake. Harvesting can occur in, or be spurred by, scheduled team meetings, calls with patent counsel, proverbial water cooler discussions, and an employee’s daily commute, such as during a solitary eureka moment.

Impediments to invention harvesting include limited bandwidth and resources, both for innovators and IP teams.

Not surprisingly, innovators’ top priority usually is to deliver technical and commercialization outcomes; projects and roles are structured and budgeted with little room for attention to IP matters or extracurricular idea generation. Similarly, IP staffing and departmental budgets typically run lean, with substantial resources being consumed by bread-and-butter IP filings, clearance work and management of defensive disputes.

Additionally, it’s challenging to plan and execute harvesting efforts, especially when multiple participants are involved—with different expertise, personalities and ways of problem solving and collaborating—and when leaders feel extreme pressure to capture consequential IP.

Invention harvesting is both a science and an art, akin to exploring uncharted realms in search of treasure. While you may have high hopes, it’s important to manage expectations: The pursuit of innovations, including investing many resources toward such pursuit, cannot assure your success. Nonetheless, if you don’t proactively embark upon exploration, you risk overlooking highly valuable technology that could bring significant competitive advantage to your enterprise or your clients.

How AI Can Facilitate Invention Harvesting

AI and related software tools can be employed in numerous ways, such as to help manage invention harvesting efforts, prepare for and execute sessions and initiatives, vet potential innovations and file for patent protection.

Such tools can improve invention harvesting outcomes, enhance communication and collaboration and reduce resource commitments of individuals and groups.

Consider the below use cases, including example vendor tools. Where anticipated usage involves proprietary or other sensitive information, organizations should consider deploying private LLMs (Large Language Models), private AI, or other appropriate solutions.

1. Scoping a harvesting session

Carefully scoping an invention harvesting session helps to ensure the right focus on the right topics that could lead to valuable innovations. It also helps paint a clearer picture of the present landscape against which innovations are to be sought.

Without such care, organizers and participants risk pursuing low-value inventions, diluting ideation due to overbroad or ill-defined objectives and covering previously-tread technical ground. In so doing, they may sabotage a current session, as well as lose organizational momentum for future efforts; meager results may breed disappointment and skepticism.

AI-enabled research and analytics tools can aid in scoping, such as in:

• Conducting landscape searches. Assessing the relevant IP, technology and commercial landscape, including status and trends, can help identify existing technologies that may be adapted or built upon in novel ways, as well as crowded and white-space areas presenting barriers and opportunities. With such awareness, harvesting efforts can be channeled appropriately. Landscaping may be especially instructive for larger-scale harvesting efforts, such as sessions striving to generate multiple solutions in one or more areas; research can support identifying and downselecting overarching session topics and providing informational context for participants.
• Assessing patentability of proposed ideas. This may be useful where an inventor already has provided an informal or formal invention disclosure. With the benefit of patentability insights, in-house or outside IP counsel can have more productive discussions with the inventor in preparation for drafting a patent application. Alternatively, such insights may call into question the value of pursuing patent protection, helping the enterprise avoid imprudent and costly decisions.

Example tools include Derwent Innovation, IP Copilot, IPRally, PatentScout, Patsnap, and TechDiscovery.

2. Creating content for communications

Coordinating invention harvesting sessions and programs may entail crafting a host of content.

Content can include informational or promotional announcements, emails, or videos to educate and motivate the target audience. For example, a communications campaign may buttress an organization-wide invention challenge in which technologists and non-technologists will be invited to propose solutions to pervasive problems in their industry.

For specific harvesting initiatives or ongoing use, internal websites or portals may be built for sharing news and updates, recognizing participant contributions, and the like.

Detailed instructions may be drafted for participants’ use before, during, and after harvesting sessions. Multiple sets of instructions may be needed, such as for facilitators participants and program champions.

Organizers can employ generative AI tools to help create and refine the above-noted communications and many others. As such, they can create higher-quality content in less time, elevate participant engagement and improve the quality of harvesting sessions.

Example tools include Canva AI, ChatGPT, Claude, Gemini, Grammarly, Meta AI, and Microsoft 365 Copilot.

3. Facilitating a session

Invention harvesting sessions may be freewheeling, even feeling chaotic, and may be fast- or slow-paced depending on format, topics, chemistry of participants, and other factors.

With many participants, lines of inquiry, groups and subgroups, physical and virtual whiteboards, Post-itÒ Notes of many colors, detours and meanderings, it can be daunting to track and document what’s unfolding in real time, to make sure key points don’t fall through the cracks and to summarize what emerged over the course of a session.

AI tools can rise to the occasion, performing transcription and summarization tasks that even the best-listening, most-efficient humans struggle with given the nature of brainstorming sessions.

Analytics tools also can be used on demand during sessions to gather intelligence to fill in knowledge gaps or preliminarily assess novelty of offered solutions. Accordingly, facilitators can influence the course of ideation, helping to ensure that creative minds in the room concentrate on promising avenues, not dead ends.

Example tools include IQ Ideas+, Lucidspark, Notta, and Otter.ai, in addition to analytics tools mentioned above.

4. Collaborating across the ecosystem

Invention harvesting fundamentally is a team sport. As a baseline, the players commonly include both innovators and IP professionals.

Innovators may be persons hired to invent in the traditional sense, such as engineers and scientists, as well as anyone invited to contribute to ideation. IP professionals may include in-house team members, outside IP counsel, IP service providers, or a combination thereof, who directly or indirectly support harvesting efforts.

The harvesting ecosystem also may include executive sponsors, other enterprise stakeholders, and bystanders with whom the process or results are shared.

Collaboration between and among the players includes in-person or remote interactions and dialogue, communication of information and provision of support related to IP and legal matters and software infrastructure, whether before, during, or after harvesting sessions.

Tools leveraging AI directly, or indirectly through complementary tools, can enable smoother, faster and more meaningful collaboration across the harvesting ecosystem.

For instance, they can enable the creation of automated or semi-automated workflows, communication and information exchange platforms, and dashboards for managing ideas and follow-ups and for reporting session results.

Improved collaboration translates into stronger outcomes and a more satisfying experience for all players involved. Beyond invention harvesting, such collaboration contributes to a vibrant IP culture in which individuals and groups are committed to innovation.

Example tools include Microsoft Forms, Pipefy, Power Automate, Power BI, Slack, and Zapier, and other tools mentioned in this article that have embedded collaboration features or interface with other relevant software.

5. Preparing invention disclosures

A high-quality invention disclosure tends to support accelerated, more accurate patentability, technology and commercial assessments and the drafting of stronger patent applications.

Unfortunately, the preparation of invention disclosures is a perennial impediment to invention harvesting.

This is understandable because conventional disclosure forms may feel cumbersome to complete, distracting inventors from their day jobs. Likewise, when IP professionals attempt to prepare disclosures on behalf of inventors, they may miss details known to the inventors or lack sufficient time to do a disclosure justice.

For large harvesting sessions that produce many ideas for consideration, problems are compounded. Time may be lost waiting for disclosures to be completed, and a highly manual process may be needed to classify and assess them. If the disclosures are incomplete or inaccurate, the quality of corresponding assessments may suffer.

AI-based tools for generating invention disclosures automatically or semi-automatically remove a key impediment to invention harvesting individually and in bulk.

By performing tasks otherwise performed by inventors or IP professionals, they enable such persons to dedicate available bandwidth primarily to ideation and related strategies rather than documentation of ideas. Moreover, by accessing a vast body of knowledge not within a single person’s purview, they equip evaluators and IP professionals with richer content and context to handle downstream steps.

Example tools include IP Copilot and Tradespace.

6. Tracking inventions for near- and far-term purposes

Conceiving an invention, or capturing a prior-conceived invention, and describing it in an invention disclosure are clear initial wins in a harvesting exercise.

However, important follow-on work must be undertaken. Such work may encompass evaluating the invention, comparing it with prior art solutions or with other potential solutions proposed by team members, strengthening the invention through further ideation or detailed testing or development, and advancing it to drafting processes described below.

A savvy enterprise aims not to undervalue, overvalue, or forget about an invention, both in the short term and over time. These aims can be elusive due to the hustle-and-bustle of IP management and tactical execution, continuously evolving R&D organizations, and overwhelmingly large data sets.

Advanced IP tools can be used to manage inventions and ideas from their inception in harvesting sessions and throughout their lifecycle. For example, IP management systems or software (IPMS) can serve as repositories for invention records, including taxonomic data characterizing related technology, product, and other attributes.

AI classifiers can auto-classify inventions into appropriate categories, enabling more efficient individual and collective review on the heels of harvesting sessions. Through ongoing tracking of inventions with pertinent metadata, inventions resulting from harvesting efforts can become part of the institutional memory of an organization, influencing subsequent IP and business strategies and tactics.

Example tools include AQX, Equinox, IPfolio, and Symphony.

7. Drafting patent applications or defensive publications

Following an invention harvesting session and subsequent assessments or refinements, a decision may be made (a) to file a patent application to protect a concept, (b) to publish the concept via a defensive publication, or (c) to more thoroughly document the concept internally to support its retention as a trade secret.

AI-based tools may assist with drafting of provisional or nonprovisional patent applications, defensive publications, or detailed internal writeups.

Example tools include Deep IP, Junior, Patlytics, PowerPatent, Rowan Patents, and Solve Intelligence.

Go Out on the AI Limb—Capture the Inventive Fruit

AI and related tools are available now to support programmatic and ad hoc invention harvesting efforts. To achieve the greatest impact, an organization should consider leveraging multiple tools to remove or substantially reduce current impediments. As a first step, it can holistically assess where it currently stands and what it desires to achieve. To the extent that relevant internal bandwidth or expertise are lacking, an IP consultant may be engaged to guide and assist.

By adopting and embracing advanced tools for invention harvesting, forward-thinking enterprises and their advisors can turn aspiration into action more than ever before.

Image Source: Deposit Photos
Author: BiancoBlue
Image ID: 581938416