Leadership Strategy for Ethical Concerns with Intellectual Property
Amanda B. Pope
Leadership and Strategy Development | AI Asset Branding and Support | Website and Product Design
Within this article, you will find leadership strategies that addresses the lack of a secure generative AI process that will not infringe upon the use of other organization’s intellectual property.
Ethical Concern
Pandora’s AI box is now open. For years, we have known that AI would transform business in most industries, but adoption, while accelerating, was slow and expensive. Foundation models have changed that: pre-trained AI can easily be used almost “out of the box” for tasks that now can be automated and improved with minimum additional training.
“Generative AI further expands the scope of what can be automated, especially in administrative, marketing, and service fields. User-friendly interfaces, such as chat or voice, have lowered or eliminated the friction at adoption.” (Goehring et., al, 2022).
The availability of large amounts of training data and advances in affordable high computing power are fueling Al’s growth. Al intersects with intellectual property (IP) in many ways.
The World Intellectual Property Organization describes intellectual property as a term classified within creations of the mind and is protected by patents, copyrights, and trademarks. (2023). However, the majority of assets and data used for AI training are challenging to regulate because many images, text, and other resources can be classified as public domain, where there is no intention for an owner to copyright the material. Furthermore, no human inventor can align a patent or copyright once AI algorithms and code are generated, producing responses, creating images, and making decisions.?
Additionally, since organizations are racing to incorporate all forms of AI, looking for ways to boost productivity faster and more effectively than the competition, thus creating Robin Hood opportunities such as,
“In the U.S., large tech corporations regularly invade on lesser firms’ intellectual property (IP). Often, this has commanded large court settlements that punish bigger corporations.” (Bannister et., al, 2020). ChatGPT has recently illustrated the multiple confidentiality and intellectual property risks introduced to businesses by foundation models such as OpenAI’s GPT4. (LaCasse, 2022).
The intellectual property used to train the models to generate derivative works is not protected; the privacy and confidentiality of new prompts and training data are not assured, and the developed work (text, code, images) can’t be copyrighted.
As Malcolm Gladwell asks in What the Dog Saw: “If words belong to the person that wrote them, and copying those words is plagiarism,” does the code belong to the developer who wrote it, or does it belong to the AI engine that will process it?
Leadership Approach
AI will transform how we work. CEOs and board directors must understand how to seize the opportunities and mitigate the enhanced risks AI presents to business. First, offer AI transparent governance. New legislation around ethical uses of AI includes data privacy and governance regulations. For example, The EU AI Act will require managing AI incidents, like data security incidents. (LaCasse, 2022). The act would also create regulatory oversight for high-risk AI applications, including hiring software and medical devices. (Engler, 2022).?
In this environment, three out of four executives say it’s vital for their companies to address data privacy and AI ethics. (Goehring et., al, 2022).? However, building trustworthy AI requires significant commitments across product engineering, IT, and governance. Tools that detect bias, diverse and inclusive teams, and guidelines for AI design can help companies develop AI that will create positive change—and an AI risk Center of Excellence can help ensure no essential steps are skipped, including establishing policies for AI ethics. As stakeholders, product engineering and governance should outline a framework for the development and use of ethical AI, and these points can be a place to start:
Additionally, leadership should assume that any antiquated designs, algorithms, and tasks were developed prior without plagiarism or intellectual property rights in mind. Enterprises must iterate in their existing productivity programs with new expectations and expanded possibilities.
To produce a continuing learning environment that engages individuals and the organization, “leaders must be successful at creating direction, alignment, and commitment.” (Hughes et al., 2014).
Furthermore, leadership must strengthen core competencies specific to the challenge and introduce a zero-trust security strategy, where a culture of modern security practices and automated controls is developed. In the event of a breach, this type of security posture helps organizations contain risks, limiting the likelihood of a material loss.
Stakeholder Analysis
Business organizations have internal and external stakeholders that directly impact business operations. Familiar stakeholders for an organization are employees, investors, customers, and suppliers. An organization cannot function without stakeholders. Stakeholders are considered the backbone of every business organization. Each stakeholder has different duties, responsibilities, and roles in business.
“Stakeholders typically value a leadership team that chooses the ethical way to accomplish the company’s legitimate for-profit goals.” (Stanberry, 2018). Organizational decisions impact stakeholders.
How much each stakeholder (for this ethical dilemma) will be affected as follows:
The ethical problem presented in this scenario involved identifying and evaluating the decision’s impact on the different stakeholders. By assessing the level of impact as low, moderate, or high for each stakeholder, the ethical decision-maker could prioritize the stakeholders and work towards meeting their moral desires.
For further debate, the employees (developers, architects, product designers) must agree that any invention, development, concept, enhancement, process, method, improvement, or any other creation becomes the organization’s exclusive property. Relates to an organization’s business process, created by an employee, is the exclusive property of the organization, and has the exclusive right to copyright. Suppose a developer or architect fails to agree to provide all details related to the intellectual property. In that case, the organization can conduct other actions to reasonably acquire the intellectual property and perfect and enforce its proprietary rights.
Operational Plan
First, implement secure, AI-first intelligent workflows to run the enterprise with generative AI. Relaunch AI-first enterprise automation programs. Change the enterprise mindset from “adding AI” to “starting with AI,” reinventing processes, tasks, workflows, and jobs to improve productivity. Reevaluate prior automation scope based on the new generative AI capabilities. Redefine jobs and skills based on the higher-value-added tasks where AI is less useful.
Next, operationalize AI and algorithmic accountability governance to design and operate trustworthy technology. Ensure use cases are easily explainable, AI-generated artifacts are identified, and AI training is transparent and open to continual critique. To manage risk, document—with fact sheets—every instance of AI use in the organization and its current governance. Ensure AI-generated assets are traceable to the foundation model, dataset, and prompt (or other inputs). Be prepared to make adjustments based on regulation changes. Re-skill the employee base to understand AI and its proper and improper use.
Then, build AI ethics and bias identification training programs for employees and partners to comply with AI ethics regulations. Accelerate the transition to zero trust across the enterprise and partner network. (Michel, 2022). Developers must avoid other conduct that is intended to or has the effect of misleading the public, funding entities, or customers about the position of the AI. This morality has an objective nature, where acting with integrity becomes the standard or baseline.
领英推荐
Organizations should position their employees to exude integrity with support and direction that includes the “need to be attentive not only on complying with the letter of the law but also on going above and beyond that basic required condition to consider their stakeholders and do what is right.” (Stanberry, 2018).
In his book Mere Christianity, C.S. Lewis suggests, “Even in literature and art, no man who bothers about originality will ever be original: whereas if you simply try to tell the truth you will, nine times out of ten, become original without ever having noticed it.” (2017).
Finally, implement AI-enabled security intelligence and ensure clear incident escalation policies are documented at every level, including the board of directors. Establish role-based controls for access to data. Implement multifactor authentication (MFA) for critical apps and data assets. (WIPO, 2023). Establish a process where research sources are noted and created, while an author and co-author (stakeholder + developer) who have contributed significantly to the concept, design, execution, or interpretation are named to the functionality.
While keeping in mind the timeliness of requesting patents, as “patent applications in the artificial intelligence (AI) field increased by 718% between 2016 and 2022, and the AI market is expected to grow to USD 191 billion by 2024.” (WIPO, 2023).??
Key IP Ethical Considerations
Intellectual property (IP) is a foremost driver of tech modernization. The current IP system was designed to foster human innovation and creation. However, as AI develops, it changes the human element of invention and opens the doors for theft and dishonesty.
Sadly, “Patent theft is a rational strategy for corporate behemoths seeking to entrench their supremacy and counterbalance a threat from a developing startup. After all, it’s cheaper to steal than to license smaller firms’ technologies.” (Michel, 2022).
With that in mind, leadership and policymakers will need to consider the following:
From a policy perspective, it is worth asking whether IP law should continue to require that humans be named as the inventor, whether it should allow an AI inventor to be named, or whether there are alternative solutions. Each possible approach to AI inventorship has potential implications across the complex IP legal framework. (Michel, 2022). Options will need to be considered in the context of local innovation ecosystems. Policymakers must keep a close eye on who the inventor is under patent law and the technical capability of AI technology to assess when these scenarios may become relevant.
Considerably, Alternative solutions to solely human inventorship or allowing AI inventorship might be a compromise. Such alternative solutions include completely removing the requirement to name an inventor, calling the person with the closest connection to the AI as the inventor, and naming a human inventor but requiring a note to be added explaining the involvement of AI. Conversely, trade secrets are an alternative to patents.
“Inventors of AI innovations (including AI models and algorithms) are faced with the dilemma of trying to patent their inventions or turning to trade secrets.” (Michel, 2022). While patents provide an easier means of enforcement, the patentability of AI innovations can be uncertain, making trade secrets a cheaper option.
Summary
Companies have long anticipated that AI would change everything one day; that day has finally arrived. Organizations are now sprinting to integrate all forms of AI, looking for ways to increase productivity faster and more efficiently than the competition. However, security, privacy, and intellectual property rights must remain paramount. In reality, people will do what’s easy more often than they do what’s right. Foundation models and generative AI are significant advancements for AI, but the immense prospect is how important it is to embed technology into the ways people work and live in the easiest-to-use way possible if we want to change what people do. (Goehring et., al, 2022). This is a critical insight for enterprise transformation and modern change management.
More forward-thinking organizations digitally have the advantage of undergoing firsthand the waves of an unsettling technology, whether in their product, service offerings, or business methods. They may also discover they need to pay more attention to morals as a matter of both practicality and necessity: They are more at risk if an issue or ethical breach arises and, given the relative pervasiveness of technology across their operations, a higher likelihood that related ethical questions will arise.
Due to the convenience of vast amounts of data and advances in inexpensive high computing power, Al’s growth. Al traverses with intellectual property (IP) in many ways. Many times, transgression is done unintentionally. To avoid being litigated for violation of usage of intellectual property, ensure there is no usage of copyrighted or trademarked material, and the company’s brand or logo is not too similar to that of others. A logo or name alone can rationally mislead somebody to think it is the other product. Also, do a patent exploration to ensure that any concepts and designs are singularly new and original. If the idea or method is similar to a product or service, receive approval to use their creative license through the proper channels.
When developers create code or build AI training, that becomes the company’s property, not the person delivering the work. Various types of IP include code, documentation, customer data, hardware, software, and many more. Correspondingly, ownership rights can be shifted to other groups or people—consequences for intellectual property violations range from fines to prison sentences. In the final recommendations, while patents provide an easier means of enforcement, the patentability of AI innovations can be uncertain, making trade secrets a cheaper option. Lastly, develop AI-enabled security acumen and ensure clear incident escalation policies are documented at every level, including the board of directors. Establish role-based controls for access to data and create reports or dashboards to monitor who is accessing the data and how it is being used.?
?
References
Bannister, C., Buckley, N., & Sniderman, B. (2020). Making ethics a priority in today’s digital organization. Deloitte Insights. https://www2.deloitte.com/us/en/insights/topics/digital-transformation/make-ethical-technology-a-priority.html
Engler, A. (2022). The EU and the U.S. are starting to align on AI regulation. Brookings. https://www.brookings.edu/blog/techtank/2022/02/01/the-eu-and-u-s-are-starting-to-align-on-ai-regulation
Gladwell, M. (2009). What the dog saw. Allen Lane.
Goehring, B., Rossi, F., & Rudden, F. (2022). AI ethics in action: An enterprise guide to progressing trustworthy AI. IBM Institute for Business Value. https://ibm.co/ai-ethics-action
Hughes, R. L., Beatty, K. C., & Dinwoodie, D. L. (2014). Becoming a strategic leader: Your role in your organization’s enduring success. Jossey-Bass.
LaCasse, A. (2022). Proposed EU AI Act blurs lines between AI developers and data processors under GDPR. Iapp. https://iapp.org/news/a/proposed-eu-ai-act-blurs-lines-between-ai-developers-and-data-processors-under-gdpr
Lewis, C. S. (2017). Mere Christianity. William Collins.
Michel, P. (2022). Big Tech has a patent violation problem. Harvard Business Review. https://hbr.org/2022/08/big-tech-has-a-patent-violation-problem
Stanberry, S.M.B. K. (2018). Business Ethics. OpenStax. https://slingshot.vitalsource.com/books/9781947172579
WIPO. (2023). What is intellectual property (IP)? World Intellectual Property Organization. https://www.wipo.int/about-ip/en/