Tinkering with AI Empowers Creativity, Innovation and Collaboration

Tinkering with AI is the Way to Build AI Capability

Recently I wrote a piece on addressing “AI hesitancy” by giving ourselves permission to put aside fear, uncertainty and doubt and have a play with these new technologies.

Here’s something else I’ve been working on to extend that idea – it’s a framework explicitly teaching an approach to capability building based on giving educators and others the confidence and permission to start tinkering with AI, solve their own problems, and create interesting things:

1. Discover and Empathise (Personal Exploration)

This stage encourages individuals to explore their own perceptions and experiences with AI and technology, to identify their concerns, needs, and challenges.

It’s about understanding one’s own relationship with technology and pinpointing areas of discomfort or interest.

2. Define (Identifying Personal Goals)

At this stage, individuals identify their own goals and the challenges they want to overcome with AI.

This could be anything from wanting to understand how AI works, to specific applications of AI they’d like to explore, or even particular problems they’d like to solve using AI.

3. Ideate (Exploring AI Possibilities)

During the ideation stage, individuals brainstorm how AI can be used to achieve their identified goals.

They explore the potential of AI and consider a broad range of possible applications. This stage encourages creativity and divergent thinking.

4. Plan (Personal Project Planning)

Next, individuals turn their chosen idea into a personal project. They plan how they will explore AI to achieve their goal.

This stage may involve outlining learning resources, creating a project timeline, identifying required tools, etc.

5. Prototype and Tinker (Hands-on AI Exploration)

This is where individuals start their hands-on exploration of AI. They begin tinkering with AI tools, learning through trial and error, creating prototypes, and not being afraid to fail and try again.

They are encouraged to learn at their own pace, exploring AI in a way that’s fun and exciting.

6. Test, Reflect, & Iterate (Learning and Improving)

Individuals test their prototypes, reflect on the process, and make changes based on what they’ve learned.

This iterative cycle allows for continuous learning and improvement. It also encourages individuals to regularly reflect on their progress and how their understanding of AI is evolving.

7. Implement & Scale (Sharing and Expanding)

At this stage, individuals take their AI projects beyond the prototyping phase. They apply their AI solutions to real-world contexts, share their work with others, and seek feedback.

They may also consider ways to expand or adapt their project based on their experiences and feedback received.

8. Review and Learn (Personal Growth)

Finally, individuals conduct a review of their journey, assessing their growth in understanding AI, the challenges they faced, the success of their projects, and what they can improve for future explorations.

This reflection helps consolidate learning and sets the stage for future tinkering and problem-solving with AI.

This process maintains a strong focus on individual empowerment, learning, and creativity. It encourages educators and others to dive into AI, giving them the freedom and confidence to explore, tinker, and create their own innovative solutions.

Where Did This Model Come From?

This model is also the result of tinkering with AI. I asked Chat GPT-4 to summarise what I thought were some of the key models and frameworks that I thought were relevant. This included the Action Research paradigm, the Innovation Cycle, Agile and a some others.

Then, once I had set that as the context, I got GPT to consolidate them into a single framework and suggest ideas for a visual. Here’s the exact conversation I had with GPT below.

Using Chat GPT to summarise, consolidate and synthesise frameworks and models

ME

Are you familiar with the action research model?

GPT

Yes, I’m familiar with the action research model. Action research is a participatory, democratic process concerned with developing practical knowledge in the pursuit of worthwhile human purposes. It is characterized by its collaborative nature and the active involvement of practitioners in the research process.

The action research model typically involves a cycle of four key steps:

1. Identifying a Problem: The initial stage involves recognizing a problem or issue in a specific context, often related to professional practice. The researchers articulate this problem clearly and determine the goals of their action research project.
2. Planning: Once the problem is identified, the next step is to devise an action plan to address it. This involves researching and considering various solutions, and then determining which is most likely to be effective.
3. Action: The planned action is then implemented. Researchers, who are often the practitioners themselves, carry out the intervention or change in their practice.
4. Reflection and Evaluation: After the action is taken, the next step is to reflect on its impact and evaluate its effectiveness. This involves analyzing data collected during the action, interpreting the results, and drawing conclusions.

These four steps are typically repeated in cycles, with the reflection and evaluation phase informing the identification of a new or refined problem, leading to a new cycle of planning, action, and evaluation. This iterative process allows for continuous learning and improvement.

It’s important to note that action research is often used in fields like education and social work where practitioners are seeking to improve their own practices and understandings, and it emphasizes the generation of practical knowledge that can inform action in specific situations.

ME

Yes, that’s right. And are you familiar with the innovation cycle?

GPT

Yes, the innovation cycle is a model that describes the process of innovation from conception to implementation. While there are many variations, a typical innovation cycle might include the following stages:

1. Ideation: This is the initial phase of the innovation cycle where new ideas are generated. It often involves brainstorming sessions, market research, and gathering insights from different stakeholders including employees, customers, or partners.
2. Concept Development: After generating a pool of ideas, the most promising ones are further developed. This might involve defining the concept more clearly, identifying potential features and benefits, and outlining how it will meet customer or market needs.
3. Feasibility Analysis or Prototyping: This stage involves testing the practicality of the idea or concept. In a product development context, this might involve developing a prototype and testing it. In a service or process context, it might involve piloting the new process or service. Financial, technical, and operational feasibility is also considered at this stage.
4. Product Development: If the idea is deemed feasible, it enters the product development phase where it’s turned into a market-ready product or service. This could involve refining the design, establishing manufacturing processes, or training staff.
5. Testing and Validation: The product, service, or process is then tested to validate that it works as expected and meets the identified needs or market demand. This could involve beta testing with a small group of customers, or it could involve large scale trials or pilots.
6. Implementation or Launch: If the testing and validation stage is successful, the product or service is launched to the market. This involves marketing and sales activities, and may also involve scaling up manufacturing or service delivery capacity.
7. Evaluation and Improvement: After the product is launched, it’s evaluated to see if it met the initial objectives and expectations. Customer feedback and performance data are collected and used to make any necessary improvements. This feedback can also inform the ideation phase for the next cycle of innovation.

The innovation cycle is designed to be iterative, with the learnings from each cycle feeding into the next, driving continuous improvement and ongoing innovation. Please note that the exact phases and their names can vary based on different models of the innovation cycle.

ME

Excellent. That’s it…! Now summarise the agile approach?

GPT

Sure! Agile is a project management and product development approach that is characterized by its flexibility, collaboration, and customer-centricity. It emphasizes iterative progress, team collaboration, and delivering a functional product to the customer in a timely manner. Here’s a summary of its key principles:

1. Iterative Development: Agile methodology breaks down projects into small, manageable units called iterations or sprints. Each sprint results in a usable product or feature, which is then tested and reviewed. Feedback from these reviews informs the direction of the following sprints.
2. Incremental Delivery: Rather than delivering a final product at the end of the project, Agile focuses on incremental delivery. After each sprint, a usable product or a part of the product is delivered to the customer. This allows for early and continuous delivery of value.
3. Collaboration and Self-organizing Teams: Agile encourages close collaboration between cross-functional teams. Teams are generally self-organizing, meaning they decide themselves how to best accomplish their tasks. Daily meetings, or stand-ups, help keep everyone informed of the project’s progress and any obstacles encountered.
4. Adaptive Planning: Agile accepts that customer needs might change over the course of a project. Therefore, instead of a detailed long-term plan, Agile uses adaptive planning, with goals and requirements often re-evaluated and adjusted at the end of each sprint based on feedback and changing needs.
5. Continuous Improvement: Agile methodology emphasizes a process of continuous refinement and improvement. After each sprint, teams reflect on what worked and what didn’t during a retrospective meeting, and they look for ways to improve in the next sprint.
6. Customer Involvement: In Agile, the customer (or customer representative) is closely involved throughout the process. They provide feedback at the end of each sprint which guides the direction of the subsequent work. This ensures that the product is closely aligned with customer needs and expectations.
7. Working Software Over Comprehensive Documentation: Agile values working software more than detailed documentation. The focus is on delivering a functioning product rather than getting caught up in comprehensive documentation. Documentation isn’t ignored but is kept as lean as necessary.

Remember, Agile is more of a mindset or philosophy than a strict process. The specific practices can vary widely based on the specific Agile framework used (like Scrum, Kanban, etc.), the nature of the project, and the organization’s culture.

ME

Perfect…! Now summarise the design thinking process or paradigm

GPT

Design thinking is a human-centered, iterative process that designers use to tackle complex problems. It involves empathy, experimentation, and rapid prototyping to innovate and create solutions that meet user needs. Here is a summary of the five main stages of the design thinking process, as popularized by Stanford’s d.school:

1. Empathize: The first step is understanding the needs of the people you’re designing for. This involves observing, interacting, and empathizing with the users to understand their experiences and motivations, as well as immersing yourself in their physical environment to gain a deeper personal understanding of the issues they face.
2. Define: In the Define stage, you’ll analyze your observations and synthesize them to define the core problems you have identified so far. This is usually expressed as a problem statement in a human-centered manner.
3. Ideate: Next, you’re ready to start generating ideas. The goal is to come up with as many solutions or innovative ideas for the problem at hand. This is where you’ll start to look for alternative ways to view the problem and identify innovative solutions.
4. Prototype: This stage is experimental. The aim is to identify the best possible solution for the problems identified during the first three stages. The solutions are implemented within the prototypes, and, one by one, they are investigated and either accepted, improved and re-examined, or rejected based on the users’ experiences.
5. Test: The final stage involves rigorous testing of the complete product using the best solutions identified during the prototyping phase. It’s an iterative stage where the results generated during the testing phase are used to redefine one or more problems and inform the understanding of the users, the conditions of use, and the product itself.

The design thinking process is iterative, flexible, and focused on collaboration between designers and users, with an emphasis on bringing ideas to life based on how real users think, feel, and behave. It’s important to remember that these stages are not always sequential — they do not have to follow any specific order and they can often occur in parallel and be repeated iteratively. As such, the design thinking model is a non-linear path that promotes a problem-solving mindset.

ME

Wonderful! Now give me a definition of “tinkering”

GPT

Tinkering refers to the act of experimenting with, modifying, repairing, or improving something in an informal, hands-on, or experimental manner. It often involves messing around with the components of a system, device, or concept to understand how it works, make it better, or repurpose it. Tinkering is often exploratory and driven by curiosity, and it can lead to both learning and innovation.

While traditionally associated with physical objects like machinery or electronics, the concept of tinkering can also be applied to abstract things like software, systems, or ideas. In the context of programming, for example, tinkering might involve playing around with different pieces of code to understand how they work or to create something new.

In education, tinkering has been recognized as a powerful way of learning, as it encourages hands-on exploration, experimentation, and learning from failure. Tinkering can foster creativity, problem-solving skills, and a deeper understanding of the subject matter.

ME

Give me a short process for a tinkering methodology, approach or mindset

GPT

Sure! Tinkering is a less structured, more exploratory process than many other methodologies, but it can still be thought of in terms of several key steps. Here’s a simplified way to conceptualize the tinkering process:

1. Curiosity and Exploration: Tinkering begins with curiosity about how something works or how it could be improved or repurposed. You might start by exploring a device, system, or concept, learning as much as you can about its components and how they interact.
2. Experimentation: Once you’ve explored the object or concept, start playing around with it. If it’s a physical object, take it apart and put it back together. If it’s a piece of software, experiment with changing parts of the code. If it’s an idea or a process, try out different variations and see what happens.
3. Making and Prototyping: Begin to make changes or create something new based on your experimentation. This could involve building a prototype of a new device, writing a new piece of software, or implementing a new process.
4. Testing and Observing: Test your creation to see how it performs. Carefully observe what happens, making note of any unexpected results.
5. Iterating and Refining: Use your observations to refine your creation. Make changes based on what you’ve learned and then test it again. Repeat this process as many times as necessary until you’re satisfied with your result.
6. Sharing and Collaborating: Share your findings, creations, or improvements with others. Collaboration is often a key part of the tinkering process, as it can bring in fresh perspectives and new ideas.

Remember, tinkering is all about hands-on exploration, learning through doing, and not being afraid to make mistakes. It’s often less about achieving a specific outcome and more about the process of experimenting, learning, and innovating. It requires a mindset of curiosity, resilience, and creativity.

ME

Now can you draw up a chart that summarises these different approaches and highlights where they are similar or different

GPT

As an AI developed by OpenAI, I’m currently unable to directly create visual content or charts. However, I can certainly describe how such a chart might look and what it might include.

The chart would likely be a table with the approaches as row labels and key aspects as column labels. Here’s a simplified text-based version:

Key Aspect / Approach	Action Research	Innovation Cycle	Agile	Design Thinking	Tinkering
Focus	Solving practical problems in specific contexts	Turning ideas into new products or services	Rapid, incremental delivery of high-quality work	Creating human-centred solutions to problems	Hands-on exploration and experimentation
Key Activities	Identifying problems, planning, taking action, reflecting & evaluating	Ideation, concept develop-ment, feasibility analysis, develop-ment, testing, launch, evaluation	Iterative develop-ment, incremental delivery, adaptive planning, continuous improve-ment	Empathizing, defining, ideating, prototyping, testing	Exploring, experimenting, making, testing, iterating, sharing
Involvement of Others	Collabora-tive; practitioners are actively involved	Often involves various stakeholders at different stages	Close team collabora-tion; customer feedback crucial	Deeply involves users; collaborative	Can be solitary but often benefits from collabora-tion
Flexibility	Iterative; cycles of action and reflection	Has structure, but feedback can influence future stages	Highly flexible; adapts to changing needs	Iterative and non-linear; stages can occur in parallel	Highly flexible; driven by curiosity and learning
Outcome Oriented	Yes, aimed at solving the identified problem	Yes, aimed at launching successful innovation	Yes, aimed at delivering usable product increments	Yes, aimed at solving user problems	Learning-oriented; any useful outcome is a bonus

I hope this helps! Remember that this is a simplified chart, and each of these approaches has a lot more depth and nuance than can be fully captured in such a format.

ME

Now I want you to design a new approach that combines all of this together. For context, I want an approach to use with educators who are terrified of AI and technology, but that we can use to introduce AI, Chat GPT and other associated technology in a way that – on the one hand, feels like “tinkering” but on the other hand incorporates the best of methodologies, approaches and concepts from Action research, the innovation cycle, agile, design thinking and tinkering

Now can you draw up a chart that summarises these different approaches and highlights where they are similar or different

GPT

Absolutely! Here’s a proposed approach that weaves together elements of action research, the innovation cycle, agile, design thinking, and tinkering:

1. Discover and Empathize (Incorporating Design Thinking & Action Research):

Start with understanding educators’ concerns and fears about AI and technology. Conduct interviews, surveys, or observations to gather their insights and learn about their needs and challenges. This aligns with the empathize stage of design thinking and also the initial problem identification phase of action research.

2. Define (Incorporating Design Thinking & Action Research):

Synthesize the information gathered to identify key barriers and define the problem(s) to be solved – for instance, “educators are afraid of AI because they fear it may replace their jobs.” This is the Define stage in design thinking, and it closely relates to defining the problem in action research.

3. Ideate (Incorporating Design Thinking & Innovation Cycle):

Brainstorm a wide range of possible solutions to the identified problems, encouraging a broad, divergent thinking process. For instance, ideas could range from developing an introductory workshop on AI to creating a community of practice where educators can learn from their peers. This corresponds to the Ideate stage in design thinking and the ideation phase of the innovation cycle.

4. Plan (Incorporating Action Research & Agile):

After selecting a few promising ideas, plan out how you will implement these solutions. Make a roadmap that breaks down the implementation process into small, manageable parts. This incorporates elements from both action research (the planning phase) and agile methodologies (adaptive planning).

5. Prototype and Tinker (Incorporating Design Thinking & Tinkering):

Develop a prototype of the proposed solution (this could be a trial workshop, a beta version of a technology tool, etc.). Then, introduce this prototype to a small group of educators and let them tinker with it. Encourage them to play around, ask questions, and make suggestions. This stage combines the prototyping phase of design thinking with the hands-on, experimental nature of tinkering.

6. Test, Reflect, & Iterate (Incorporating Agile, Design Thinking, & Action Research):

Observe and collect feedback as educators interact with the prototype. Reflect on the feedback, and test the solution again after making necessary modifications. This phase should be a repeating cycle, allowing for continuous learning and improvement – much like the iterative development in agile, the test phase in design thinking, and the reflection and evaluation in action research.

7. Implement & Scale (Incorporating Innovation Cycle & Agile):

Once the solution has been refined and tested sufficiently, roll it out to a wider audience of educators. Remember to keep the process agile – continue to seek feedback, make improvements, and adapt to changing needs even after the full implementation. This corresponds to the launch phase of the innovation cycle and the incremental delivery principle of agile.

8. Review and Learn (Incorporating all methods):

Finally, conduct a formal review of the whole project. What worked? What didn’t? What can be improved for next time? How have educators’ attitudes towards AI and technology changed? Document these learnings and use them to inform future initiatives.

This new approach is designed to be flexible, iterative, and user-centered, with an emphasis on hands-on learning and experimentation – hence, it should be particularly suitable for educators who are apprehensive about AI and technology.

ME

Give me ideas for a graphic I can use to represent this visually

GPT

Absolutely! Here’s a description for a graphic that could represent this proposed approach:

1. Start with a Large Circle: This circle represents the overall process and the cyclical, iterative nature of the approach.
2. Divide the Circle into 8 Segments: Each segment represents one of the eight stages in the approach. Label each segment with the corresponding stage name: “Discover and Empathize,” “Define,” “Ideate,” “Plan,” “Prototype and Tinker,” “Test, Reflect, & Iterate,” “Implement & Scale,” and “Review and Learn.”
3. Icons and Colors: Assign a unique color and a simple icon to each stage to make the graphic more engaging and easier to understand. For example:

• Discover and Empathize: An eye (for observation) or two people talking (for interviews)
• Define: A lightbulb (for idea)
• Ideate: A cloud of thought bubbles or a brainstorm icon
• Plan: A roadmap or checklist
• Prototype and Tinker: A hand manipulating gears or tools
• Test, Reflect, & Iterate: A magnifying glass examining a product or an arrow forming a small loop (representing iteration)
• Implement & Scale: An upward-pointing arrow or a group of people (representing wider audience)
• Review and Learn: A book or a graduation cap (symbolizing learning)

1. Inner Circle: Inside the larger circle, create a smaller circle where you can add labels or icons representing the various methodologies being incorporated: action research, the innovation cycle, agile, design thinking, and tinkering. This will show that all these approaches are central to the process.
2. Arrows: Show arrows moving from one stage to the next to highlight the process flow. You could also have a curved arrow going from the “Review and Learn” stage back to “Discover and Empathize” to emphasize the cyclical, iterative nature of the process.
3. Outer Annotations: Outside the circle, you might want to add brief descriptions of each stage or key activities that take place in each stage.

Remember, the aim of the graphic is to visually communicate the process in a clear, engaging way. The specific design elements can be adjusted to best fit your audience and context.

ME

Suggest some names for this approach

GPT

Sure! Here are some suggestions that capture the essence of this holistic and iterative approach:

1. Integrated Innovation Approach (IIA): This name emphasizes the way the approach integrates multiple methodologies to foster innovation in education.
2. Iterative Empowerment Methodology (IEM): This name highlights the iterative, cyclical nature of the process, and its goal of empowering educators through hands-on learning and experimentation.
3. Holistic Educator Engagement (HEE): This name underscores the approach’s focus on engaging educators in a comprehensive and immersive learning experience.
4. Agile Educator Empowerment Process (AEEP): This name highlights the agile and adaptive nature of the process, and its focus on empowering educators.
5. Educator-centric Innovation Cycle (EIC): This name emphasizes the user-centered, educator-focused nature of the approach, and its resemblance to a cycle of innovation.
6. Learning through Tinkering and Collaboration (LTC): This name underscores the approach’s emphasis on hands-on learning (tinkering) and collaborative problem-solving.
7. Tinkering Toward Tech Comfort (T3C): This name reflects the process’s focus on easing educators into comfort with AI and other technologies through a hands-on, tinkering approach.

Remember, the best name will depend on the exact context and audience, and it should capture the unique aspects and goals of the approach in a catchy, easy-to-remember way.

ME

Could you give me a couple of translations for this into te reo Māori? They don’t have to be exactly literal word for word

GPT

Sure, I can provide translations for some of the proposed names into te reo Māori, the indigenous language of the Māori people in New Zealand. Please note, te reo Māori has its unique cultural nuances and literal translation might not always encapsulate the essence of the original English term. Here are a few attempts:

1. Integrated Innovation Approach (IIA): “Te Ara Auaha Whakakotahi”
2. Iterative Empowerment Methodology (IEM): “Te Rarangi Whakamana Takirua”
3. Holistic Educator Engagement (HEE): “Whai Wāhi Katoa a te Kaiwhakaako”

Please note that you should ideally consult with a fluent te reo Māori speaker or professional translator to ensure the accuracy and appropriateness of these translations.

ME

For this one: Learning through Tinkering and Collaboration (LTC)

GPT

In te reo Māori, the verb “whakatikatika” is typically used to mean “to correct” or “to set right”. However, in the context of the phrase “Te Akoranga Mā Te Whakatikatika Me Te Mahi Tahi”, we’re using it to represent the concept of “tinkering”, which involves making adjustments or modifications, often in an experimental or hands-on way. So, while “whakatikatika” doesn’t directly translate to “tinkering”, it’s being used here to convey a similar concept.

As always, it’s important to consult with a fluent speaker or professional translator to ensure the most accurate and culturally appropriate translation. Language is deeply tied to culture and context, and this is especially true for te reo Māori.

If you found this interesting or useful

Check out my other content on AI in Education here. There’s more on how you can use Chat GPT to enhance learning.