Using AI to Learn About AI: A Meta-Learning Journey
- X4 Consulting
- 5 days ago
- 4 min read
At X4 Consulting, we work with organisations to discover how technology can help businesses and make a real difference.
Our team are curious and like to learn by sharing knowledge, doing research, and applying this through working with clients. Over the last few months we tried something new: using AI to learn more about AI and using that knowledge to upskill in AI tools. Our learning is being applied and we’re adapting as we go.
The Importance of AI Literacy for New Zealand Businesses
AI literacy means more than just understanding what the technology is and how it works. It’s about knowing what is available and any limitations so you can choose the right AI tools for your business problem.
Implementation of any AI tool, or tool set, must be in line with ethics, privacy, and security considerations. Increasing literacy of AI tools supports a gradual introduction of AI should be following a systematic approach.
Selecting the Right AI for Your Needs
With countless AI tools available, knowing which tool fits your specific challenge means you can harness AI effectively to support, innovation, and problem-solving. Balance being realistic and grounded with curiosity to:
Confirm real use cases: Know what AI can truly do for you, have clear and specific business specific challenges.
Know constraints: Check outputs to notice problems like hallucinations or accuracy issues.
Test for the right fit: Pick solutions based on results, not marketing or hype.
Getting Started with AI in New Zealand
New Zealand has excellent AI resources to help you begin your journey. These include:
The AI Forum of New Zealand - A non-profit bringing together New Zealand's AI community with reports, guidance, and tools for different audiences.
Government Chief Digital Officer (GCDO) - Supporting safe AI adoption across government with practical guidance.
Additional resources:
NZ AI Policy Tracker (Brainbox Institute)
Each offers different types of content—from high-level strategy documents to practical implementation guides.
Our Meta-Learning Experiment: AI as Teacher
Our consultants were already using AI in daily life—trip planning, research, curating playlists, and managing complex meal planning. When we decided to upskill formally, we wondered: could we use AI tools as learning partners, not just subjects of study?
We experimented with ChatGPT, Gemini, Claude, and Copilot as both topics and teachers. Our consultants used these tools to:
Research and recommend free relevant courses
Build personalised learning plans suited to individual learning styles
Explore mailing lists, communities and news feeds on AI
Test prompts exploring topics like natural language processing, ethics, and generative AI
Practice hands-on applications to reinforce concepts
Explore specific prompting techniques for real-world scenarios
Document learning iterations and progress
What We Discovered
Using AI to learn about AI revealed both opportunities and challenges:
The wins: You can get started quickly with a draft you can build on, personalised learning experiences that you can expand on, significant time savings in research, near-instantaneous feedback, and increased confidence through hands-on practice.
The challenges: Choosing the right tool for each task (no single "Swiss Army knife" exists), managing AI's tendency toward overly agreeable responses when constructive feedback was needed, and navigating occasionally outdated information. Being realistic and managing your own expectations on what AI can do based on the training data.
Key insights: Keep a curious open mind. AI works best as a collaborator, not a replacement. Developing strong prompting skills is practical and valuable, though it requires patience and iteration.
Our biggest takeaways
Don’t be afraid to experiment, use free versions - Many AI tools have free versions. This is a great way to familiarise yourself with the tools before considering whether they are going to meet your needs and justify investment.
Check your privacy settings - Check your account settings to see what the security, privacy and data control or protection default settings are. Initially the information you’re using to become familiar with how the tools work shouldnt be sensitive and should never contain information that might be subject to copyright. Check with your organisation to be sure you understand any internal policies or standards on use of AI tools.
Start small - There are so many different types of AI tools out there, we’re almost spoilt for choice. Start with exploring and using tools that are relevant to the tasks you’re likely to want AI to support you with.
Mix and match - You may find the best approach that works for some tasks is a multi-tool approach, this could address the individual limitations each tool has when used alone.
Be prepared for frustration - Regular use quickly reveals where AI fails, hallucinates, or produces inconsistent results, helping you understand realistic boundaries. Outputs can sound confident but double checking or asking for sources helps identify inaccuracies.
Practice - Small prompt changes can produce quite different results. AI results can be shaped or altered depending on how you shape or change your input.
It’s a conversation, not a query - Prompting is initiating a conversation, rather than crafting a search query, it’s a dialogue with AI — you set the context and guide the response, shaping it along the way.
What's Next for AI at X4?
We've recently updated our AI Code of Ethics to guide our thoughtful adoption of new technology. Our approach is anchored by principles including:
Transparency and accountability
Human-AI collaboration
Privacy and data protection
Government alignment
Aotearoa-specific considerations
Continuous improvement through feedback
As we continue exploring AI's potential, we remain committed to responsible implementation that delivers genuine value for our clients and the broader New Zealand business community.
Want to explore how AI could support your organisation's goals? Get in touch to discuss your specific challenges and opportunities.
Full transparency: We used AI to help structure this blog post and refine some wording—practicing what we preach about human-AI collaboration.
