Navigating the Ethics of Large Language Models in HCI Research: Opportunities and Challenges

Core Concepts

HCI researchers are increasingly integrating large language models (LLMs) across their research workflows, from ideation to system development and paper writing. While researchers acknowledge a range of ethical concerns, they often struggle to identify and address those issues in their own projects.

Abstract

The study examines how HCI researchers are using large language models (LLMs) across various stages of their research process, and the ethical considerations they encounter.
Key highlights:

HCI researchers use LLMs for ideation, literature review, study design, data analysis, system building and evaluation, and paper writing. They perceive LLMs as opening up new possibilities for their research.
Researchers express a range of ethical concerns, including:

Harms of engaging with LLM outputs (e.g., biased, discriminatory, or fabricated information)
Threats to privacy of participant data
Violations of intellectual integrity (e.g., ambiguity around ownership of LLM-generated content)
Overtrust and overreliance on LLMs
Environmental and societal impacts

However, researchers often struggle to identify and address these ethical concerns in their own projects. They employ strategies like conditional and reactive engagement with ethics, limited disclosure practices, restricting LLM use, and delaying responsibility.
The findings highlight the need for the HCI community to proactively engage with research ethics around LLM use, through mechanisms like IRB involvement, developing ethical frameworks, and shifting academic incentives.

Stats

"LLMs are useful but not creative" (P15)
"we have all these models that are competing against each other, that's like millions of dollars of electricity and computer components, [which is] really bad for the environment" (P2)
"I think many people are very hasty to say yes or no. And I think that's not the answer. The answer is always in a gray area." (P2)
"To me, it is not gonna be helpful in researchers assessing the credibility or validity of the work. It is just like a meta issue about how the actual document was formed and refined." (P10)

Quotes

"the main problem is that I don't know what bias it has, and I don't know how to figure it out." (P16)
"the challenge is the regulation needs to come much later rather than early. If you come up with regulations too early, you may kill innovation and on the other hand, you don't know what you want to regulate because the representation of the product hasn't been stabilized yet." (P12)
"LLMs are still an unknown territory, so people don't know how to react, I assume." (P9)

Key Insights Distilled From

"I'm categorizing LLM as a productivity tool"

by Shivani Kapa... at arxiv.org 04-01-2024

https://arxiv.org/pdf/2403.19876.pdf

Deeper Inquiries

How can the HCI community develop effective guidelines and frameworks to support responsible use of LLMs in research, while also fostering innovation?

In order to develop effective guidelines and frameworks for the responsible use of Large Language Models (LLMs) in Human-Computer Interaction (HCI) research, the HCI community can take several steps:

Collaborative Efforts: The HCI community can collaborate with experts in AI ethics, data privacy, and responsible AI to develop comprehensive guidelines that address the unique ethical considerations of using LLMs in research.

Ethics Education: Incorporating ethics education into HCI research programs can help researchers understand the implications of using LLMs and equip them with the knowledge to navigate ethical challenges effectively.

Transparency and Disclosure: Encouraging researchers to be transparent about their use of LLMs in research publications and presentations can promote accountability and trust within the community.

Ethics Review Boards: Establishing specialized ethics review boards within HCI institutions can provide oversight and guidance on the ethical implications of using LLMs in research projects.

Continuous Evaluation: Regularly evaluating the impact of LLM use on research outcomes and participant experiences can help identify and address ethical concerns in a timely manner.

Community Dialogues: Hosting workshops, panels, and discussions within the HCI community can facilitate conversations about the responsible use of LLMs and foster a culture of ethical awareness and accountability.

By implementing these strategies, the HCI community can develop robust guidelines and frameworks that promote the responsible use of LLMs in research while also fostering innovation and creativity in the field.

How can the potential unintended consequences of HCI researchers' current strategies of restricting LLM use and delaying responsibility be mitigated?

The potential unintended consequences of HCI researchers' current strategies of restricting LLM use and delaying responsibility can be mitigated through the following approaches:

Proactive Ethical Considerations: Encouraging researchers to proactively identify and address ethical concerns related to LLM use at the outset of their research projects can help prevent unintended consequences.

Training and Education: Providing researchers with training on ethical decision-making and responsible AI practices can empower them to navigate ethical challenges effectively and make informed decisions.

Collaborative Decision-Making: Promoting collaborative decision-making processes within research teams can ensure that ethical considerations are discussed and addressed collectively, reducing the risk of unintended consequences.

Regular Monitoring and Evaluation: Implementing mechanisms for regular monitoring and evaluation of LLM use in research projects can help researchers identify and mitigate any unintended consequences in a timely manner.

Transparency and Accountability: Emphasizing transparency in LLM use and accountability for ethical decision-making can help researchers take ownership of the potential consequences of their actions and work towards mitigating them.

By adopting these strategies, HCI researchers can minimize the potential unintended consequences of restricting LLM use and delaying responsibility, promoting ethical research practices and positive outcomes in their projects.

Given the distributed nature of responsibility in the LLM supply chain, how can HCI researchers, LLM providers, and other stakeholders collaborate to address ethical concerns around LLM use in research?

Collaboration among HCI researchers, LLM providers, and other stakeholders is essential to address ethical concerns around LLM use in research. Here are some ways they can collaborate effectively:

Establish Clear Communication Channels: Creating open lines of communication between HCI researchers and LLM providers can facilitate the sharing of information on ethical considerations, data privacy, and model capabilities.

Joint Ethical Guidelines: Collaboratively developing ethical guidelines and frameworks that outline best practices for using LLMs in research can ensure that all stakeholders are aligned on ethical standards and responsibilities.

Regular Consultations and Reviews: Conducting regular consultations and reviews between HCI researchers and LLM providers can help identify and address ethical concerns in a timely manner, fostering a culture of ethical awareness and accountability.

Training and Education Programs: Offering training and education programs for researchers, LLM providers, and other stakeholders on ethical AI practices and responsible data use can promote a shared understanding of ethical considerations and best practices.

Ethics Review Boards: Establishing joint ethics review boards that include representatives from HCI research, LLM providers, and other relevant stakeholders can provide oversight and guidance on the ethical implications of using LLMs in research projects.

By collaborating closely and engaging in ongoing dialogue, HCI researchers, LLM providers, and other stakeholders can work together to address ethical concerns around LLM use in research, promoting responsible and ethical practices in the field.