How AI Agents Can Reduce Agency in Data Analysis (Full Transcript)

Using AI for coding and interpretation can boost speed but risks skill erosion. Learn how to retain control and evaluate AI output quality.
Download Transcript (DOCX)
Speakers
add Add new speaker

[00:00:00] Speaker 1: The agency here, as you can see, is moderate. It's a little bit of concern because you are losing your control. But when you use AI model, see what happens here. You are offloading a lot of what you have to do for the AI to do it for you. Maybe you're asking the AI to suggest code for you, to develop things for you, to write the interpretation, to produce a report. All these are causing you to depend on the AI system in making sense of your data, and your agency is very low. It's low to moderate. You don't have much control over the data analysis process, and you are losing your skill. That's a concern. The ability to code your data, analyze your data, is going down because now AI is doing that for you. We are in a stage of AI agent, where you can ask AI to do a task for you, and it will independently make decisions, and then bring you your outcome. In this situation, let's say you want to analyze your qualitative data, and you have an AI agent. What you could do is to give all the transfer to the AI, and there are three things that you can give to the AI. The background information about a study, the objective about a study, and also the outcome that you want. And you leave everything to AI too, and AI can go make decisions, and then within minutes, you get your result. So you have offloaded a lot of important decisions to AI to make. So the cognitive offload is very high when you use an agent, and your agency is low now. You don't have control over the decisions that the system makes for you. And as you are losing control, you're also losing the skill and knowledge, right? So that is the problem that we are going to face in this kind of situation. I think you have to be aware, and in my presentation, I'm going to talk about what actions that you can take in order to not lose your agency, right? To be able to assess the quality of the information the AI too produces, right? And that's what I'm going to really talk about. I hope this introduction helps you. This is just a brief information. These are the skills that you need. The good thing about this dashboard is that there are some scenarios that you can learn from, right? And we can click on it and get all the information that you need. And also the sales assessment is also here, just to help you to know the level of skill that you have in this kind of situation.

ai AI Insights
Arow Summary
Speaker explains how using AI models—and especially autonomous AI agents—can reduce a researcher’s agency in qualitative data analysis by offloading coding, interpretation, and reporting decisions to the system. This high cognitive offload may erode users’ skills and control over the analytic process. The speaker previews a presentation focused on actions to retain agency and assess AI output quality, and references a dashboard with scenarios and skills assessment to support learning.
Arow Title
AI, Cognitive Offload, and Losing Agency in Data Analysis
Arow Keywords
agency Remove
cognitive offload Remove
AI models Remove
AI agents Remove
qualitative data analysis Remove
coding skills Remove
decision-making Remove
control Remove
skill erosion Remove
assessment dashboard Remove
AI literacy Remove
output quality evaluation Remove
Arow Key Takeaways
  • Greater AI assistance can lower user agency by shifting key analytic decisions to the system.
  • Autonomous AI agents increase cognitive offload by independently making decisions and returning results quickly.
  • Reliance on AI for coding, interpretation, and reporting can erode researchers’ analysis skills over time.
  • Users should develop practices to maintain control and critically assess the quality of AI-produced information.
  • Scenario-based dashboards and skills assessments can help learners understand risks and build competencies.
Arow Sentiments
Neutral: The tone is cautionary but informational, emphasizing concerns about loss of control and skills while outlining planned guidance and learning resources.
Arow Enter your query
{{ secondsToHumanTime(time) }}
Back
Forward
{{ Math.round(speed * 100) / 100 }}x
{{ secondsToHumanTime(duration) }}
close
New speaker
Add speaker
close
Edit speaker
Save changes
close
Share Transcript