H - Technologies

Demonstrate proficiency in identifying, using, and evaluating current and emerging information and communication technologies.

Introduction and Explication

“The public library was conceived in an age of information scarcity, while today’s networked world is one of information abundance and mobility” (Garmer, 2014). Digital Information Age information organizations embrace this abundance and technological mobility by providing access to communications and information technologies; implementing user-centered policies and procedures; and developing safe, innovative services which meet the information needs of information seekers and reflect global and community-based socio-economic trends.

Identifying, Using, and Evaluating the Technology Landscape

Current information and communications technologies are constantly evolving and new technologies are emerging rapidly. For example, the Internet of Things enables users to engage, participate, and share knowledge with others at the global scale (Figueroa, 2014; Hirsh, 2018 p.3). Artificial intelligence (AI) systems also offer great benefits to information communities, as does open data and open access to digital information systems, repositories, and resources. From a library and information science (LIS) perspective, this enhanced knowledge sharing capacity reflects our professional mandate to provide broad and equitable access and use of data and information, and, as such we embrace them and advocate for their use in our organizations (Hirsh, 2018, p. 3).

Digital Information Age information professionals recognize that current and emerging information and communications technologies provide our constituent communities with unprecedented access to information and knowledge. As such, we embrace these tools, advocate for their strategic integration into our digital and physical organizational environments; introduce relevant innovative services; "design [user-centered] creative spaces and instructional tools and create new opportunities for learning and exploration" (Hirsh, 2018, p. 3). Examples of such services include but are not limited to online access to collections (e.g., digital libraries); online reference services (e.g., ChatGPT), and physical innovation labs supportive of entrepreneurial activities and knowledge creation and sharing (Hirsh, 2018, p. 7). We also recognize the risks to intellectual freedom and other human rights created by the complex machine learning algorithms driving socio-technical systems such as artificial intelligence (AI) and, particularly, machine learning algorithms.

Because machine learning relies on large sets of user data to generate relevant responses to user queries, lifecycle documentation of the design, implementation, use, and evaluation is a critical element of trustworthy technology management (Jo and Gebru, 2020). Further, informed information professionals recognize that some forms of artificial intelligence pose great risks to data privacy due to their opacity, reliance on big data, and their capacity to perpetuate biased and discriminatory practices that misrepresent and/or marginalize vulnerable peoples from participating in, and benefitting from, this technologically-situated knowledge sharing endeavor.

Educating Users

Today's information organizations are responsible for proactively identifying evolving and emerging communications and information technologies, strategically integrating them into our service structures and workflows, and encouraging broad, equitable, and informed user engagement with them. Therefore, safe, user-centric digital literacy instruction has become a core responsibility of the effective Digital Information Age library and information science (LIS) professional.

Information professionals uphold the principle of democratic inclusivity in, and long-term sustainability for, information organizations by encouraging informed user engagement with current and emerging technologies. We assume a strategic human-rights centered, user-directed approach to the deployment of technologies and related services in our organizations; teach literacy to our community members; educate ourselves and our stakeholders about the societal risks and benefits of these technologies; implement trustworthy policies and procedures for their design, implementation, and use; and transparently and proactively strive to protect all stakeholders' personally identifiable data and information (Hirsh, 2018). Moreover, in keeping with our professional mandate, we call out and directly intervene in the Digital Divide by ensuring that all our community members have convenient access to, use of, and instruction for, these socio-technological resources.

Evidence

Evidence 1:     Archives & Recommender Systems Links to an external site.                   

Archives & Recommender Systems is a short paper written for the seminar, Archives and AI (MARA 284, S22) in which I address the benefits and risks to human rights posed by machine learning algorithms an apply them to an archival context. The rapid design and deployment of socio-technical systems at a global scale have the potential to bring great benefits to society including the capacity to enhance individual and collective well-being (AI HLEG, 2019). As a form of AI, recommender systems are machine learning algorithms which aggregate, analyze, and use data to identify and promote resources to users in response to their system queries (Mooradian, 2019). Recommender systems (and all machine learning algorithms) are trained on large sets of user data which may result in, and perpetuate, biased and discriminatory system decision-making. Further, the opacity of machine learning systems may negatively impact their interpretability, explainability, transparency, and contestability by users (Mooradian, 2019). Examples of such harmful system practices include user profiling and reducing users' ease of access by declining to promote less popular information resources (Jo & Gebru, 2019; Milano et al, 2019).

To reduce risks to stakeholders, digital archives should assume a human-centered approach by applying content-based filtering to machine learning-enhanced collections processing workflows. As opposed to collaborative filtering which relies on user data to make decisions, content-based filtering concentrates on collection objects' standards-compliant metadata to identify accurate user recommendations (Arnold, 2016). Significantly, Arnold (2017) suggests that an added benefit of content-filtering recommender systems is the creation of new links between previously disparate digital objects which may “produce a new structural argument and provide a narrative that differs from other organization systems within a digital archive," that can "guide users to unexplored parts of a large collection of items” (Arnold et al, 2017). As an example, the National Archives and Records Administration (NARA) released a new user interface in 2022 which employs content-filtering machine learning algorithms to identify and retrieve relevant information in response to user queries. To educate users about how to protect their personal data when interacting with the new system, NARA includes numerous interactive multi-media resources and protective options on the interface.

This paper represents my knowledge and skills as relates to Competency H by providing a guide to archives seeking to deploy recommender systems which, like all AI systems, require human-oversight and governance if they are to meet their potential to contribute to societal well-being, just information access and use, and participatory knowledge creation and sharing. By articulating the risks and benefits of information and communications technologies within the context of user needs and trends, the paper elucidates my commitment to the values and principles of the LIS domain, and offer actionable, usable, and meaningful strategies for effective Digital Information Age stakeholder advocacy, education, and data privacy protection.  

Evidence 2:     Capstone Project: Trustworthy Generative AI Links to an external site.     

Trustworthy Generative AI is my capstone project for AI, Data, and Ethics (INFO 297, F23). The paper introduces ethical benefits and risks to society posed by all AI systems as identified by the European Union's high-level expert group on artificial intelligence (AI HLEG, 2019). Within this context, I present a hypothetical scenario in which an organization is seeking to develop a governance program founded on the principles of trustworthy AI, and which both facilitates the use of ChatGPT as a marketing strategy and ensures organizational compliance with data privacy and protection laws and standards.

ChatGPT (and other generative bots) learns from training data and user attributes to generate answers to user queries (Jo, 2022). At this time, however, these systems are not designed with data privacy guarantees, nor is it possible to remove individual data from training sets or predict how a chatbot will reuse data (Daniels, 2023). Therefore, it is essential that organizations seeking to expand their customer engagement, via ChatGPT or another artificial intelligence assistant, consider the voluntary AI ethics landscape, comply with relevant laws and regulations, and craft an information governance program within their organizations that incorporates the AI HLEG (2019) principles for trustworthy AI into their design, use, and evaluation of these socio-technical systems. As part of this governance program, organizational policies and procedures should ensure the transparency of data collection policies, practices, and intended uses; provide interpretable and understandable explanations of system processes in support of individual control of private data; and overtly acknowledge that ChatGPT's default setting automatically opts users into all functional choices; and obtain informed consent from all users.

Strategic human-centric, principles-based, information and knowledge governance lies at the heart of the effective information professional's overarching mandate and daily practice. Further, the LIS profession's ethical values and best practices align with those outlined by AI HLEG (2019) which include human oversight and agency; technical robustness and safety; privacy and data governance; transparency; diversity, non-discrimination, and fairness; accountability; and societal and environmental well-being (AI HLEG, 2019). Therefore, this assessment strategy acts as evidence of my knowledge and skills in Competency H by outlining a values-based approach to developing a trustworthy AI governance program by which an information organization may craft, oversee, and iteratively assess their use of an interactive artificial intelligence assistant to expand and improve customer engagement and organizational efficiency.

Evidence 3:     Silicon Kids of the '70s Presentation Links to an external site.

Silicon Kids of the '70s is a video presentation I made in MARA 284, Community Archives (Su21). It provides an overview of my written proposal for a digital inclusive community memory project/archives that I designed to collect, preserve, represent, celebrate, and make broadly accessible, the memories of all children living in 1970s’ Silicon Valley, with a particular focus on those voices missing from the existing narrative. The project strives to build a sustainable, standards-informed, archival collection of these memories, and to preserve them in an online, digital repository for broad and equitable representation and access to researchers, and to assure long term preservation of the collection so that these voices are heard and maintained as part of the comprehensive historical record.

This presentation represents my knowledge and skills related to Competency H by elucidating my ability to utilize information and communications technologies to create a fully online, globally-accessible participatory project using Microsoft 365, Google Workspace, WordPress, SoundCloud, audio recording apps, and numerous social media platforms. Additionally, it incorporates a detailed strategy for the digital repository's design including the use of open, OAIS-compliant archival software for long term preservation of and access to participant recordings, as well as data privacy and stakeholder confidentiality policies. A project budget and funding strategies are also addressed.

Conclusion

To meet their professional mandate to provide safe, broad, and equitable access to and use of knowledge and information, Digital Information Age information professionals must maintain proficiency in identifying, using, and evaluating current and emerging information and communications technologies. While socio-technical systems such as AI have the capacity to bring great benefit to society, they are not currently designed to protect data privacy rights. Therefore, as knowledge facilitators in an increasingly connected and participatory world, information professionals are responsible for providing users with opportunities to learn and use these technologies in a safe and informed manner, and to understand the risks they pose to their data privacy and the strategies they may employ to protect themselves from data misuse or reuse.

References

Arnold, T. (2016). Recommender systems for digital archives. IPAM. http://www.ipam.ucla.edu/abstract/?tid=13210&pcode=CAWS2 Links to an external site.

Arnold, Leonard, P., & Tilton, L. (2017). Knowledge creation through recommender systems. Digital Scholarship in the Humanities, 32(suppl_2), ii151–ii157. https://doi.org/10.1093/llc/fqx035 Links to an external site.

AI HLEG. (2019). Ethics guidelines for trustworthy AI. Shaping Europe’s Digital Future. https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai Links to an external site.

Daniels, J. (2023, May 1). How generative AI can affect your business’ data privacy. Forbes. https://www.forbes.com/sites/forbesbusinesscouncil/2023/05/01/how-generative-ai-can-affect-your-business-data-privacy/?sh=793797a9702d Links to an external site.

Figueroa, M. (2014). Internet of Things. Tools, Publications & Resources. https://www.ala.org/tools/future/trends/IoT Links to an external site.

Garmer, A. (2014, October 14). Rising to the challenge: Re-envisioning public libraries. The Aspen Institute. https://www.aspeninstitute.org/publications/rising-challenge-re-envisioning-public-libraries/ Links to an external site.

Hirsh, S. (2018). Information services today: an introduction (2nd ed.). Rowman & Littlefield.

Jo, H. (2022). Impact of information security on continuance intention of artificial intelligence assistant. Procedia Computer Science, 204, 768–774. Science Direct. https://doi.org/10.1016/j.procs.2022.08.093 Links to an external site.

Jo, E. S., & Gebru, T. (2020). Lessons from archives. Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency, 306–316. https://doi.org/10.1145/3351095.3372829 Links to an external site.

Milano, S., Taddeo, M., & Floridi, L. (2019). Recommender systems and their ethical challenges. SSRN Electronic Journal, 35(4), 957–967. https://doi.org/10.2139/ssrn.3378581 Links to an external site.

Mooradian, N.A. (2019). AI, records, and accountability. Information Management Journal, 53(5), 9–13.