In my ongoing journey to transform my newsletter into a digital garden, I discovered an unlikely collaborator: Generative AI. At first glance, the idea of using an AI as a “critical friend” may seem unconventional, even paradoxical. After all, how could an algorithm provide the thoughtful, nuanced critique we often expect from a trusted confidant? Yet, approaching Generative AI with a beginner’s mindset revealed its potential as a valuable partner in organizing, revising, and refining my ideas.
In this post, I’ll explore what it means to use AI as a critical friend, share examples of our conversations, and reflect on how this collaboration has reshaped my thinking. In most of my use cases listed below, I used ChatGPT, but I’ll explore other tools and uses in upcoming posts.
What Does It Mean to Use AI as a Critical Friend?
A critical friend is someone who supports your growth by providing constructive feedback, asking thoughtful questions, and challenging assumptions. Traditionally, this role is filled by a trusted colleague, mentor, or collaborator—someone who brings an outside perspective while understanding your goals.
When I invited ChatGPT to step into this role, I treated it not as a tool to dictate answers but as a sounding board. I asked open-ended questions, sought clarification on my ideas, and encouraged iterative refinement of content. ChatGPT’s strengths lie in its ability to:
- Provide a beginner’s perspective, helping me identify gaps in clarity or coherence.
- Generate alternative ways of framing ideas or structuring content.
- Offer a consistent and readily available partner for brainstorming.
While it lacks the lived experience and emotional intelligence of a human critical friend, its ability to rapidly process and reorganize ideas proved invaluable for certain tasks.
A Human Parallel: Lessons from My Dissertation
As an example of this type of collaborative process, I’m reminded of the work I did with my father while writing my dissertation. We regularly met via video conferencing tools and spent hours reading through sections, unpacking my thinking, cleaning up my writing, and trying to weed out jargon. My father, an attorney by trade, knew how to write but wasn’t from my field. His perspective helped me become more explicit in my explanations, ensuring my work was accessible to readers outside my niche.
This experience mirrors what I’ve found with Generative AI—though not human; it offers a fresh perspective that helps me step back from my work and see it through new eyes. Unlike my father, ChatGPT lacks emotional and lived context, but it similarly encourages clarity and challenges assumptions, albeit in its own way.
Pros and Cons of Using ChatGPT as a Critical Friend
Pros
- Debugging Ideas: ChatGPT can help pinpoint where my ideas or writing break down, offering suggestions to refine them.
- Beginner’s Mindset: It approaches problems with a fresh perspective, often identifying areas that might confuse a first-time reader.
- Broad Knowledge Base: It knows things I don’t, such as Python coding techniques or obscure historical facts, which can be helpful in unexpected ways.
Cons
- Limited Perspective: ChatGPT can’t see what I see or fully grasp the context of my work. I found myself typing out descriptions of web pages or content that I couldn’t copy/paste from my screen.
- Short-Term Memory Issues: While it remembers the current conversation, it forgets decisions we’ve made in previous sessions.
- Quick Fix Mentality: It often prioritizes solving problems quickly rather than exploring deeper, more nuanced solutions.
- Unreliable Coding Suggestions: When using it for coding, ChatGPT sometimes suggests overly complex solutions or ideas that might break my work. I need a middling understanding of what I’m doing to assess its suggestions effectively.
These challenges remind me that while AI can enhance creativity, it also requires careful oversight and thoughtful integration.
Examples of Collaboration with ChatGPT
Here are some concrete examples of how ChatGPT and I worked together to refine my newsletter content for the digital garden:
1. Clarifying Concepts
One of my early struggles was articulating the purpose of my digital garden and its distinction from my newsletter. I asked ChatGPT:
Prompt: How would you explain the difference between a newsletter and a digital garden to someone unfamiliar with either?
Response: A newsletter is a curated collection of insights sent periodically to subscribers, often focused on timely or thematic updates. A digital garden, in contrast, is a living, interconnected repository of knowledge that evolves over time, encouraging exploration and reflection.
This succinct comparison became a foundation for how I communicated the transition to my audience.
2. Structuring Content
Organizing 400 newsletter issues into a cohesive digital garden felt overwhelming. I asked ChatGPT for advice:
Prompt: What are some ways I can categorize and link newsletter content in a digital garden?
Response: Consider thematic categories such as “Productivity,” “Digital Literacy,” and “Personal Growth.” Use tags and crosslinks to connect related content. Adding a ‘Map of Content’ page can provide an overview for readers to navigate the garden easily.
These suggestions helped me draft an initial structure, which I refined further.
3. Generating Reflective Prompts
To make my digital garden more interactive, I wanted to include reflection questions for readers. ChatGPT’s input here was particularly helpful:
Prompt: Can you suggest reflection questions related to digital literacy?
Response: 1. How do you define digital literacy in your own life? 2. What tools or practices help you stay organized and informed online? 3. In what ways can you foster digital resilience in your community?
These prompts became part of the “Reflect and Engage” section in many of my revised posts.
Reflections on AI as a Critical Friend
Strengths
- Beginner’s Mindset: ChatGPT’s lack of preconceptions often led to fresh perspectives. It identified areas where my explanations might confuse a reader unfamiliar with the topic.
- Speed and Consistency: Unlike a human collaborator, ChatGPT was always available to brainstorm or iterate on ideas, no matter the time of day.
- Encouraging Iteration: By providing quick feedback, ChatGPT enabled me to test multiple drafts and ideas without feeling bogged down by the process.
Weaknesses
- Lack of Depth: While useful for surface-level critique and organization, ChatGPT struggled with deeply nuanced or context-specific insights.
- Overconfidence: At times, its responses appeared authoritative but lacked accuracy or subtlety, requiring careful vetting on my part.
- Forgetting Decisions: ChatGPT often forgets the context or decisions made in prior sessions, leading to repetitive work.
Looking Ahead
This experiment with ChatGPT as a critical friend has reshaped how I view collaboration. It’s not a replacement for human insight but a complementary tool that can enhance productivity and creativity when used thoughtfully. In upcoming posts, I’ll dive deeper into the strengths and weaknesses of generative AI in projects like this, exploring how to balance its contributions with the irreplaceable value of human collaboration.
For now, I’d love to hear from you: How do you approach collaboration with AI? Have you ever treated an AI as a critical friend? Let’s discuss this in the comments or connect directly. Together, we can redefine what collaboration looks like in the digital age.
Photo by Ayush Kumar on Unsplash