Patent Clearance Strategies for Ai Startups: Conducting an Effective FTO Analysis

In the current era, blooming of artificial intelligence (AI) is revolutionary. AI- driven technologies like financial forecasting, AI powered healthcare, innovations in biotechnology, nano technology, renewable energy technology, IoT driven sustainability and large language learning models (LLM) like GPT-4, Gemini to autonomous systems are reshaping how businesses operates and communication with people and technology. The rise of generative AI has ignited evolution in software development, content creation and even in creative arts which further makes AI more accessible to everyone and it has been integrated into mainstream applications like natural language processing (NLP), computer vision and Machine learning (ML).

The rapid growth in AI initiated the evolution of AI based startups which led to create intense competition in innovation and also created a havoc for more legal battles over the intellectual property rights. As technologies are getting more advanced everyday AI based companies begin to file patents aggressively in order to protect their innovations. This has led to legal war for patents. For example, the dispute between the Google’s waymo and Uber for innovation in self-driving technology, where Uber was required to pay 245 million dollars to settle a lawsuit concerning misappropriated trade secrets. Similarly in 2023, stability AI has faced legal challenges from another company Getty images for training its AI model with their copyrighted photographs without their permission. From these examples, it is clearly understood that there is growing importance to create awareness on Freedom to operate (FTO) analysis which facilitates the AI startups to explore the complex patent landscape and avoid costly legal disputes.

Why AI Startups must Prioritize Patent Clearance 

Patent clearance becomes essential for AI based startups. Since the evolution of AI technology, unlike traditional software, innovations and inventions in AI often integrate with multiple technologies which makes them more prone to patent overlapping and risk for infringement. A single AI product might involve multiple patented components like machine learning models, data processing techniques and user interfaces which may be protected in the existing patents owned by the competitors. AI based startups risks unintentionally for infringing on third-party intellectual property rights which may lead to fines, expensive litigation and restrictions on the distribution of products. For example, in 2017, Deep mind faced scrutiny for its AI integrated health app for the usage of data rights which becomes an eye opener for many company to give importance for the thorough analysis of IP assessments. Clearing the patent directly impact the AI startups ability to attract investments from investors and venture capitalists which can lead to legal risks and uncertainty.

AI Startup
[Image Sources: Shutterstock]

A well-documented freedom to operate (FTO) analysis provides an idea for the investors about the company that has taken proactive steps to minimize risks and ensure long term viability. Startups may struggle to secure more funds or high yielding partnerships without completing FTO analysis because stakeholders may fear for the risk of litigation. As regulations of AI is given importance globally, patent compliance is becoming even more critical. A strong FTO strategy ensures that the product is compliant with regional patent laws and licensing requirements particularly for the startups which aim to operate globally. AI startups can avoid legal battles, enhance credibility and establish a competitive advantage by integrating patent clearance for their invention and innovation. Some important patent clearance strategies for AI based startups are discussed below.

Patent landscape monitoring

AI based startups should continuously monitor the patent landscape from the idea stage, thereby it is not necessary to wait until the development of a product is complete. As AI is rapidly evolving field, it is crucial to stay updated on abandoned patents, new filings or licensing opportunities.

How to Implement?

  • AI driven patent search tools plays a vital role to identify that the innovation has already been claimed. Some of the tools like Derwent innovation, Patent scout or the lens to track is very important to identify works in real time.
  • Review patent classification codes relevant to AI like G06N in regular interval of time.
  • Conducting the quarterly or biannual patent audits which is essential for the publication of new patent in the way that existing invention do not cause a great threat for FTO status.

Example:

An AI startup is developing an autonomous drone navigation software which notices that a competitor has already filed a similar patent covering path optimization method. By identifying this early, the startup can either modify the algorithm to design or explore the licensing options or update the product before the launch of the product.

Why it matters?

Monitoring the patents reduces the risk of unexpected legal disputes which also facilitates the startups to make informed decision about product development, strategic partnerships and potential acquisition based on the published patent.

Innovate smartly: Avoiding Patent Infringement

If an AI based startups identifies an existing patent document that could possibly create an infringement risk, it can create a new design around the existing patent by creating modifications in the technology. This can be done through re-engineering specific components in the existing technology or modifying or using the alternative approaches that do not overlap with patent claims.

How to Implement?

  • Focusing on the patent claims is very important. First conduct the detailed analysis of patent claims and second focus specifically on the independent claims as it describes about the invention.
  • Third, collaborate with subject matter experts and legal experts to explore the alternative methods.
  • Finally, document all the modifications and technical differentiators for the future references to publish new patent.

Example

Assume, a startup is working on AI integrated predictive analytics for healthcare, identifies that a multinational company is holding patent on deep learning based anomaly detection method, instead of infringing, the company can switch to different methods like reinforcement learning or develop a new hybrid AI model which achieves the same outcome but with different approach.

Why it matters?

Designing around the patents facilitates AI startups to maintain FTO without spending for legal battles or expensive licensing fees, thereby promoting innovation without any single barrier.

Harness Open source and Patent Pooling Strategies for growth

AI based startups can strategically use the Open source based AI models to participate in patent pools in order to gain access for the existing patented technologies and reduce the risk of legal battles. Open source licensing should be used in an appropriate manner which provides legal framework for startups to develop AI applications without violating any aspect of existing patent document.

How to Implement?

  • AI based frameworks must be utilized appropriately with permissive licenses. For example Tensor flow under Apache2.0 and Py torch under BSD.
  • Check and analyze AI patent pools and cross licensing agreement to avoid the risk of litigation. For example LOT network.
  • Strictly adhere to open source licensing terms in order to avoid unforeseen legal complications.

Example

 Google’s Tensor flow is widely used for the development of AI model because of its open source licensing model Apache 2.0 which permits modification and distributes without claiming any royalty payments. An AI startup focusing on natural language processing (NLP) can be developed on tensor flow rather than developing a new proprietary model, thereby avoiding the potential infringement.

Why it Matters?

Leveraging open source models facilitated AI based startups to reduce developmental costs, minimize legal risk and accelerate innovation by developing over the existing technologies.

Secure Licensing Agreements and Early Patent Partnerships

If a startup identifies, that technology integrated in the existing patent by another inventor is the most important aspect for their product, then instead of claiming through legal battles, they can negotiate for early licensing agreement or collaborating with the patent holder for a strategic partnership to ensure smooth market entry. Instead of considering patent as a barrier for their innovation, AI based startup companies can consider it as an opportunity for making new collaborations.

How to Implement?

  • Identify the important and specific patents which align with your AI product.
  • Approach one who holds the patent to get license early or formal agreement to commit for joint ventures and research and development collaborations.
  • Cross licensing agreement is helpful for both the parties as it is delivered through shared IP.

Example

AI based startup Wayve, focused on self-driving technology, thereby it further initiated the partnership with Microsoft to gain access for their Azure AI super computing capabilities. This strategic partnership helped wayve to leverage Microsoft’s IP when securing access to cutting edge computing infrastructure.

Why it Matters?

Ensure to receive licenses early, in order to prevent legal disputes, foster industry collaboration and allow startup companies to focus more on innovation rather than litigation.

Defensive publishing to Prevent future patent claims

Startup companies mainly focuses only on patenting their own innovations but defensive publishing is also an equally powerful strategy in which deliberately disclose the invention to the public claiming for the authority and also prevent the competitors from patenting the same invention. Defensive publishing safeguard the inventions from only restrictive patent claims which gives way to develop providing freedom to innovate.

How to Implement?

  • Publishing technical papers, whitepapers, or open source documentation through which details of the AI methods, approach, algorithm or architecture can be revealed to the public.
  • Submitting the innovations in prior art databases such as IP.com, prior art database or MIT’s technical disclosure commons.
  • Ensure publications are incorporated with detailed technical descriptions, to qualify as prior art as well as to block future patent filings on similar technology.

Example

In 2014, Tesla open-sourced its electric vehicle patents not only to encourage more innovations but also to prevent competitors from claiming key technology.

Why it Matters?

Defensive publishing blocks patent trolls by securing exclusive rights over innovations in AI. This method enables startups to work freely without fear of infringement of claims on their disclosed technology.

From Patent landscape monitoring, Innovate smartly: Avoiding Patent Infringement, Harness Open source and Patent Pooling Strategies for growth, Secure Licensing Agreements and Early Patent Partnerships and Defensive publishing to prevent future patent claims, startups have multiple avenues to ensure freedom to operate (FTO) analysis while fostering innovation.

By adopting these five strategies AI based startups can protect intellectual property rights, mitigate legal risk and establish a strong foothold in their respective domains. As AI continues to bloom in the current era, one who approach patent clearance as a strategy will lead the future of AI innovation.

Author- Dr. S.Sarayu Priyadharshini, in case of any queries please contact/write back to us via email to chhavi@khuranaandkhurana.com or at Khurana & Khurana, Advocates and IP Attorney.

References

  1. (n.d.). Revolutionizing patent risk assessment: The value and practical implications of the Evalify’s infringement by usage. Evalify AI. Retrieved from https://www.evalify.ai/blog/revolutionizing-patent-risk-assessment-the-value-and-practical-implications-of-the-evalifys-infringement-by-usage
  2. Foley & Lardner LLP. (2022, October). 5 key IP considerations for AI startups. Foley & Lardner LLP. Retrieved from https://www.foley.com/insights/publications/2022/10/5-key-ip-considerations-ai-startups/
  3. The National Law Review. (n.d.). Best practices for developing winning IP strategies for AI companies. The National Law Review. Retrieved from https://natlawreview.com/article/best-practices-developing-winning-ip-strategies-ai-companies
  4. com. (n.d.). Defensive publishing and innovation protection strategies. IP.com. Retrieved from https://ip.com/innovation-power-suite/defensive-publishing/
  5. (n.d.). Patent strategy and administration: Defensive publication. Questel. Retrieved from https://www.questel.com/patent/patent-strategy-and-administration/defensive-publication/
  6. Hern, A. (2017, May 16). Google DeepMind 1.6m patient record deal was ‘inappropriate’. The Guardian. Retrieved from https://www.theguardian.com/technology/2017/may/16/google-deepmind-16m-patient-record-deal-inappropriate-data-guardian-royal-free

Leave a Reply

Categories

Archives

  • March 2025
  • February 2025
  • January 2025
  • December 2024
  • November 2024
  • October 2024
  • September 2024
  • August 2024
  • July 2024
  • June 2024
  • May 2024
  • April 2024
  • March 2024
  • February 2024
  • January 2024
  • December 2023
  • November 2023
  • October 2023
  • September 2023
  • August 2023
  • July 2023
  • June 2023
  • May 2023
  • April 2023
  • March 2023
  • February 2023
  • January 2023
  • December 2022
  • November 2022
  • October 2022
  • September 2022
  • August 2022
  • July 2022
  • June 2022
  • May 2022
  • April 2022
  • March 2022
  • February 2022
  • January 2022
  • December 2021
  • November 2021
  • October 2021
  • September 2021
  • August 2021
  • July 2021
  • June 2021
  • May 2021
  • April 2021
  • March 2021
  • February 2021
  • January 2021
  • December 2020
  • November 2020
  • October 2020
  • September 2020
  • August 2020
  • July 2020
  • June 2020
  • May 2020
  • April 2020
  • March 2020
  • February 2020
  • January 2020
  • December 2019
  • November 2019
  • October 2019
  • September 2019
  • August 2019
  • July 2019
  • June 2019
  • May 2019
  • April 2019
  • March 2019
  • February 2019
  • January 2019
  • December 2018
  • November 2018
  • October 2018
  • September 2018
  • August 2018
  • July 2018
  • June 2018
  • May 2018
  • April 2018
  • March 2018
  • February 2018
  • January 2018
  • December 2017
  • November 2017
  • September 2017
  • August 2017
  • July 2017
  • June 2017
  • May 2017
  • April 2017
  • March 2017
  • February 2017
  • January 2017
  • December 2016
  • November 2016
  • October 2016
  • September 2016
  • August 2016
  • July 2016
  • June 2016
  • May 2016
  • April 2016
  • March 2016
  • February 2016
  • January 2016
  • December 2015
  • November 2015
  • October 2015
  • September 2015
  • August 2015
  • July 2015
  • June 2015
  • May 2015
  • April 2015
  • March 2015
  • February 2015
  • January 2015
  • December 2014
  • November 2014
  • October 2014
  • September 2014
  • August 2014
  • July 2014
  • May 2014
  • April 2014
  • March 2014
  • February 2014
  • January 2014
  • December 2013
  • November 2013
  • October 2013
  • September 2013
  • August 2013
  • July 2013
  • June 2013
  • May 2013
  • April 2013
  • March 2013
  • February 2013
  • January 2013
  • December 2012
  • November 2012
  • September 2012
  • August 2012
  • July 2012
  • June 2012
  • May 2012
  • April 2012
  • March 2012
  • February 2012
  • January 2012
  • December 2011
  • November 2011
  • October 2011
  • September 2011
  • August 2011
  • July 2011
  • June 2011
  • May 2011
  • April 2011
  • February 2011
  • January 2011
  • December 2010
  • September 2010
  • July 2010
  • June 2010
  • May 2010
  • April 2010