Ethical and Responsible AI Is Not Just a Technical Question - Reflections from DocShop

AI is transforming the way researchers search for information, analyse data, and communicate their work. Yet the most important questions surrounding AI are often not technical ones. Who benefits from these technologies? Whose knowledge is represented? And how can researchers use AI while upholding principles of integrity, responsibility, and equity? These were among the key themes discussed at the DocShop on Ethical and Responsible Approaches to AI, which brought together experts and researchers to reflect on the opportunities and challenges AI presents for academia and global knowledge production.

Artificial intelligence is rapidly becoming part of everyday academic work. Researchers use AI for example to search for literature, refine their writing, organise information, and support data analysis. The question is no longer simply whether researchers should use AI, but instead, how can researchers use AI responsibly? What happens to academic integrity in an AI-assisted world? And who benefits from these technologies and who risks being left behind?

These questions were some of the questions explored during the DocShop on Ethical and Responsible Approaches to AI, which brought together doctoral researchers, scholars, and experts from Finland and beyond to explore the promises, risks, and inequalities associated with AI in research.

AI does not change the foundations of research

The workshop opened with keynote presentation by Simo Kyllönen who presented the new recommendations of the Finnish National Board on Research Integrity (TENK), emphasizing that AI should be understood as a tool rather than a replacement for researcher themselves. Research integrity, transparency, critical thinking and accountability remain just as important in an age of AI as they were before. Researchers remain responsible for the quality, reliability, and ethical conduct of their work, regardless of which technologies they use.

In the second keynote presentation, Professor Kudzayi Tarisayi invited participants to think about what a doctoral degree ultimately recognises: when a researcher graduates, what is being celebrated? A polished document? Or the intellectual journey, judgement, and original thinking behind it?

Kudzayi Tarisayi argued that AI should serve as a partner that enhances human capabilities rather than replacing them. Researchers still need to ask critical questions, evaluate information, make decisions, and exercise scholarly judgement. As he put it, the goal is not artificial intelligence instead of human intelligence, but these two working together.

Looking at AI through a global lens

As a network committed to strengthening higher education and research partnerships between Global North and Global South, UniPID wanted to delve into how emerging technologies shape global inequalities and opportunities. The invited panelists, Ilkka Jormanainen, Maria Ndapewa Ntinda, Liisa Laakso and the two above-mentioned keynote speakers, all highlighted that access to AI is not distributed evenly. While some universities have access to advanced tools, infrastructure, and training opportunities, other continue to face challenges related to internet connectivity, computing resources and institutional support.

The discussion also challenged participants to think beyond access alone. Who creates AI systems? Whose knowledge is included in training data? Which languages are represented? And whose realities are reflected in the outputs that AI generates?

Panelists pointed out how much of the data used to train AI systems originates from the Global North. This creates risk that knowledge, languages, and perspectives from other regions become underrepresented or misunderstood. For researchers working on global development, these concerns are particularly significant. Knowledge is never neutral, and neither are the systems built upon it.

One insight that repeatedly surfaced during the discussion was the distinction between access and capacity. Having access to AI tools does not automatically mean that researchers benefit from them equally. Panelists highlighted that education will play a central role in ensuring that AI becomes a tool for inclusion rather than exclusion. From schools to universities, learners need opportunities not only to use AI but also to understand it.

Researchers should not navigate AI alone

The conversation also moved beyond individual responsibility. While researchers are expected to use AI ethically, panelists emphasized that responsibility cannot rest solely on individual researchers. Universities, research communities, funders, and policymakers all have a role to play in creating environments where responsible AI use is possible and promoted. This means providing guidance, developing clear policies, investing in secure tools, and creating opportunities to discuss challenges and dilemmas.

Events such as DocShop are one way of doing that: creating space where people can openly discuss uncertainties, learn from one another and collectively navigate a rapidly changing landscape.

Building a more equitable AI future

As the discussion turned towards the future, panelists shared their ideas about more equitable and responsible AI ecosystem. They hoped for stronger support for open-access knowledge, greater inclusion of underrepresented languages and perspectives, increased collaboration between Global North and Global South institutions, improved AI literacy, and more diverse participation in the development of AI.

Underlying these proposals was a shared belief that AI should not reproduce existing inequalities. Instead, it should help broaden participation in knowledge production and support more inclusive forms of research and learning.

The event concluded with reflections from Anders Danell who highlighted three themes: agency, responsibility and equity. These themes offer useful guiding as researchers continue to explore the role of AI in academia.

Conversations about AI are ultimately conversations about the future of research, education and global cooperation. They are opportunities to ask not only what technologies can do, but also what kind of academic community we want to build.

DocShop did not seek to provide definitive answers. Instead, it created space for reflection, dialogue and critical engagement. In a field evolving as quickly as AI, that may be one of the most valuable outcomes we can hope for.

***

Missed this DocShop? Don't worry! You can find the videorecording of the event here: UniPID & NAI DocShop on Ethical and Responsible Approaches to AI | 1 October 2026