📦 Think (Really) Outside the Box with Creative Constraints

Dialogic #334

Leadership, learning, innovation

Your Snapshot
A summary of the key insights from this issue

⬩ Constraints are not always bad for creativity, but can actually stimulate it in some cases.

⬩ Leaders and facilitators can use constraints to focus efforts and prompt more creative thinking.

⬩ The type, level, and flexibility of constraints, as well as individual factors, influence how they affect creativity.

Think (Really) Outside the Box with Creative Constraints

Many people think that creativity means thinking outside the box without any rules or limits. But is that true? Studies show that some limitations can help us be more creative by giving us focus, guidance, and challenge.

However, not all limits are good for creativity, and some may be harmful. In this issue, we explore how to use boundaries to think outside the box and how we can apply this to our projects.

What are constraints, and how do they affect creativity?

Constraints are any factors that limit your range of possible solutions or actions. They can be external (time, budget, or resources) or internal (goals, skills, or beliefs).

Constraints can affect creativity differently, depending on how we perceive and process them. Some constraints may hinder creativity by reducing motivation, increasing stress, or blocking alternative ideas.

Other constraints may enhance creativity by focusing attention, stimulating divergent thinking, or triggering novel associations.

Which constraints are more impactful for creativity?

Not all constraints have the same impact on creativity. Some research has identified five types of constraints that are relevant to creative work:

  • Time constraints refer to deadlines or schedules that limit the duration of your creative process.
  • Competition constraints refer to the awareness of rivals or competitors in your context.
  • Resource constraints refer to the availability or scarcity of materials, tools, or information needed for your creative task.
  • Anchoring constraints refer to the initial conditions or parameters that define the scope or direction of the task.
  • Autonomy constraints refer to the degree of freedom or control the individual has over the creative task.

According to a meta-analysis of 145 studies, time and competition constraints negatively affect creativity, while resource, anchoring, and autonomy constraints have positive effects.

However, these effects are not guaranteed and may differ from one situation to another.

How can leaders or facilitators use constraints effectively?

If you are a leader or facilitator, you can use constraints as a strategic tool to enhance creativity. Here are some suggestions for using constraints effectively:

  • Use constraints deliberately: Don’t avoid them entirely, but don’t impose them randomly. Choose constraints that are relevant and meaningful for the creative task at hand. For example, you might use resource or anchoring constraints to stimulate different perspectives if you want to generate new project ideas.
  • Prioritise high-impact constraints: Focus on the types of constraints that have positive effects on creativity, such as resources, anchoring, and autonomy. Avoid or minimise constraints that have adverse effects, such as time and competition. For example, suppose you want to improve the quality of creative output. In that case, you might give your team members more autonomy (agency = voice, choice and responsibility) and reduce unnecessary deadlines or pressure.
  • Allow flexibility in constraints: Ensure room for interpretation and adaptation in how constraints are applied. Don’t make them too rigid or fixed. Allow people to adjust or modify constraints as they progress in their creative work. For example, if you want to encourage experimentation and exploration, you might allow people to change the project parameters as they discover new possibilities.

Strategically using constraints can enhance creativity by providing structure, direction and challenge. Still, it requires choosing effective constraints and allowing flexibility.

⏭🎯 Your Next Steps
Commit to action and turn words into works

Identify a current or upcoming creative task you or your team are working on. List the existing or potential constraints that apply to it. Evaluate how they affect your creativity. Keep the ones that are helpful and eliminate or reduce the harmful ones.

Experiment with introducing new constraints to your creative task. Choose limitations that are relevant and meaningful for your job. Try different types of constraints such as resource, anchoring, or autonomy. Observe how they influence your creative thinking and output.

Share your experience and insights with others. Discuss how constraints have helped or hindered your creativity. Learn from others’ perspectives and practices. Seek feedback and suggestions on how to improve your use of constraints.

🗣💬 Your Talking Points
Lead a team dialogue with these provocations

⬩ What examples of constraints have enhanced or hindered your creativity in the past? How did you cope with them?

⬩ What are some types of constraints that you find more stimulating or challenging for your creativity? Why?

⬩ How do you communicate and negotiate constraints with others? How do you handle conflicts or disagreements over constraints?

🕳🐇 Down the Rabbit Hole
Still curious? Explore some further readings from my archive

Here are some of my articles about creativity and constraints.

3 Activities to Help Your Team: Generate, Develop and Judge Ideas “There are hundreds of different activities you can use for generating and developing ideas. I thought I would share a trio that works well together. They each require a specific type of thinking or mindset to be successful. The three activities are good representative examples of Divergent, Emergent and Convergent thinking.”

Imagination, Augmented: How AI Can Be a Creativity AmplifierThe future of creativity is not just human or AI, but a blend of both, offering us a unique opportunity to reach new heights of creative potential.”

Finding the edges of your page “Not to be confused with restraint which is much more about self-control, constraint is about finding the edges of the page before you begin, it is about knowing what limits you have in terms of resources.”

Thanks for reading. Drop me a note with any Kind, Specific and Helpful feedback about this issue. I always enjoy hearing from readers.

~ Tom Barrett

humAIn

A learning community for educators to connect and explore our humanity in the age of artificial intelligence.

Limited places for Australian educators.


*Applications close October 1, 2023

Support this newsletter

Donate by leaving a tip

Encourage a colleague to subscribe​​​​​

Tweet about this issue

The Bunurong people of the Kulin Nation are the Traditional Custodians of the land on which I write and create. I recognise their continuing connection and stewardship of lands, waters, communities and learning. I pay my respects to Indigenous Elders past, present and those who are emerging. Sovereignty has never been ceded. It always was and always will be Aboriginal land.

Unsubscribe | Update your profile | Mt Eliza, Melbourne, VIC 3930

.: Promptcraft 27 .: Clone your voice into 7 languages​

Hello Reader,

Welcome to Promptcraft, your weekly newsletter on artificial intelligence for education. Every Monday, I curate the latest news, tools and resources so you can consider how AI changes how we teach and learn.

In this issue:

  • ✨ NEW: humAIn learning community for educators
  • Google set to release Gemini to compete with GPT-4
  • Adobe’s Firefly generative AI tools are now widely available
  • 5,000 free audiobooks released by Microsoft and Project Gutenberg

Let’s get started!

.: Tom

Latest News

.: AI Updates & Developments

.: Microsoft and Project Gutenberg release over 5,000 free audiobooks (the-decoder.com) ➜ AI technologies were used to create over 5,000 free audiobooks with high-quality synthetic voices. The project has collected more than 35,000 hours of audio data on various genres of literature, which will be made available as open source without restrictions.

.: Stability AI debuts Stable Audio bringing text to audio generation to the masses (VentureBeat) ➜ A new technology from Stability AI that allows users to generate audio clips from text prompts, using a diffusion model trained on over 800,000 pieces of licensed music.

.: This new AI video tool clones your voice in 7 languagesHeyGen has launched a new tool that clones your voice from a video and translates what you’re saying into seven different languages. If that wasn’t enough, it also syncs your lips to your new voice so the final clip looks (and sounds) as realistic as possible.

Here’s a quick two minute demo from Jon Finger.

video preview

Pretty amazing! What do you think?

.: Adobe’s Firefly generative AI tools are now widely available ➜ Firefly-powered features like Photoshop’s Generative Fill tool are now out of beta. Adobe is also introducing a new bonus scheme to pay its stock contributors for helping to train its AI models.

bukkykvX9jGdfdE2NgdH9y

.: Content Credentials — a digital “nutrition label”By default, Firefly includes Content Credentials on every asset created using Firefly to indicate that generative AI was used, bringing more trust and transparency to digital content. Content Credentials are verifiable details that serve as a digital “nutrition label.”

.: Google set to release Gemini model soon ➜ Gemini is Google’s upcoming AI system, developed by DeepMind, that aims to compete with OpenAI’s GPT models. It is designed to be multimodal, integrating text, images, and other data types, and to have abilities like memory, planning, and reasoning.

.: EU to let ‘responsible’ AI startups train models on its supercomputers | TechCrunch ➜ EU plans to expand access to HPC supercomputers for AI startups The EU wants to support AI innovation that aligns with its values and governance framework. It will offer access to its high-performance computing infrastructure for selected AI startups to train their models faster and cheaper.

.: ‘Overwhelming consensus’ on AI regulation – Musk – BBC News ➜ Tech leaders such as Elon Musk, Mark Zuckerberg, Sundar Pichai, Bill Gates and Satya Nadella met with US lawmakers to discuss AI regulation. There was “overwhelming consensus” for some form of oversight on AI to prevent potential harms and dangers to the public.

Reflection

.: Why this news matters for education

Before moving on to other stories, it’s essential we highlight the troubling lack of diversity in the recent private meeting in the US.

This exclusive gathering failed to represent or include the perspectives of many. Let’s keep in mind the need for more inclusive representation in positions of power.

As Avril Ussery Sisk explains:

This meeting is a start, and an incomplete one at that, because the range of viewpoints and interests necessary for good policymaking is not present.

Further exploration of this here from Andrea Bonime-Blanc. And as another commenter puts it:

Those men represent my interests about as much as a space alien does.

~

Elsewhere, the demonstration of the new Heygen video translation really stuck in my mind this week. If you haven’t seen it, scroll up and play the video I shared.

It reminded me of the first time I saw the Google Lens translate from image tool – which is still pretty stunning.

Why does this type of advance matter for education? Our schools encompass diverse communities with a wide variety of languages and cultures.

Imagine if AI-powered translation tools like Heygen’s facilitate even marginal improvements in communication and understanding between these groups – say 10%, 15% or 20%. The impact on learning could be immense.

Another aspect of this is the facilitation of learning access. Imagine a time when a student’s ambition is no longer limited by a language barrier.

Or, as Matt Esterman puts it

any internet connected student in the world could “attend” any school in the world

.:

~ Tom

Prompts

.: Refine your promptcraft

Two parts to this section today:

  1. A reminder about Custom Instructions in ChatGPT
  2. Share your prompts with me and get them featured here.

Custom Instructions

I stumbled on a great article from Every this week, which reminded me to make the most of the Custom Instructions setting in ChatGPT. (We looked at Custom Instructions in issue #19.)

This allows you to enter information to tailor the interactions to your context and personalise the experience.

The author, Dan Shipper, explains:

Custom Instructions are good enough that I’d go so far as to say it is a crime if you are not using them. It’s a theft of your talent, time, and potential.

Read the full article here which includes a great breakdown of how to craft precise instructions.

~

Share your promptcraft

Are you sitting on some gems?
  • An invitation for you to send me some examples of promptcraft, which has been really effective.
  • Feel free to share screenshots or ChatGPT shareable links.
  • I will feature some reader prompts, as it is critical we continue to share our experiences and highlight good examples.
  • Hit reply or email me here >> tom@dialogiclearning.com

.:

Remember to make this your own, tinker and evaluate the completions.

Learning

.: Boost your AI Literacy

.: ✨ NEW: humAIn learning community for educators

A learning community for educators to connect and explore our humanity in the age of artificial intelligence.

We have 40 places in our October 2023 Cohort only available to Australian educators; apply before October 1 to save your place.

.: Prompts for Education: Enhancing Productivity & Learning

A prompt collection on GitHub.

By using these prompts, staff can save time and work more efficiently, and students can explore new and exciting learning opportunities. Whether you’re a student, a third-grade teacher, a college professor, or a school administrator, this collection is designed with you in mind. No technical expertise required!

.: How can we prepare students for transformed roles if we ban them from using their future tools?

An answer directed at academic supervisors (and perhaps everyone else who is banning AI tools) from Professor Marek Kowalkiewicz.

I’ll try to put it gently: come out from that rock you live under. Yes, generative AI can be a problematic technology, but that doesn’t grant you the liberty to ignore its existence. Grasp the reality and grasp generative AI’s good, bad, and ugly facets. Afterwards, offer your students informed advice. If your stance remains against generative AI, that’s fine—you are the supervisor, after all. But explain your reasoning to them. Simply put, this is your job. A blanket ban doesn’t cut it.

Ethics

.: Provocations for Balance

“As AI systems continue to advance, ask yourself: Whose voices and experiences are represented in the training data for these models? Whose are absent or excluded? Does the team building this AI reflect the full richness of humanity? Progress cannot be measured by technological prowess alone. Unless the thinkers and communities shaping AI represent the beautiful plurality of humankind, we risk building a future that reflects only a fraction of our potential.”

~ Inspired by this week’s developments.

.:

That’s all for this week; I hope you enjoyed this issue of Promptcraft. I would love some kind, specific and helpful feedback.

If you have any questions, comments, stories to share or suggestions for future topics, please reply to this email or contact me at tom@dialogiclearning.com

The more we invest in our understanding of AI, the more powerful and effective our educational systems become. Thanks for being part of our growing community!

Please pay it forward by sharing the Promptcraft signup page with your networks or colleagues.

.: Tom Barrett

/Creator /Coach /Consultant

🏃‍♂️ How a 50km Run Pushed Me to Rethink My Limits

Dialogic #333

Leadership, learning, innovation

Your Snapshot
A summary of the key insights from this issue

⬩ We tend to underestimate our full potential when setting goals for ourselves or our teams, failing to acknowledge the upper limits of our capabilities.

⬩ The cognitive bias of “illusory inferiority” leads high-achievers to set their sights too low, establishing goals below their actual potential.

⬩ Reframing our self-imposed limitations with fresh perspective can help us set and achieve more ambitious goals beyond our comfort zones.

How a 50km Run Pushed Me to Rethink My Limits

Last Saturday, I ran 50km, the longest distance I’ve ever covered in a single day. I followed it up with a 14km run on Sunday, bringing my weekly total to over 100km.

For a long time, I’ve had my sights set on surpassing 50km, so it’s exhilarating to be in this post-goal period, reflecting on what’s next.

This week’s issue is about reframing, adjusting your goals and understanding your true limits.

We’re gonna need a bigger goal

At some point in a long distance run you switch into a ‘just get it done’ disposition. Perhaps it occurs when you have already achieved the majority of the running distance, and I experienced the same switch when I ran my first marathon a few months ago.

As soon as I knew I could reach my 50km goal last weekend, I was thinking about the next goal. Which caused me to reflect on ‘what am I truly capable of?”

When setting goals for our personal growth or professional projects, we tend to underestimate what we are capable of. You might hear this referred to as the “glass ceiling” effect or even ‘setting the bar too low”.

A critical cognitive bias is at play here that is worth considering, as it will likely change how you write goals in the future.

It is common for highly competent people to underrate their abilities and establish goals below their potential. This is called illusory inferiority. You might have read about the Dunning-Kruger effect, which explains how unskilled people often rate their abilities higher than they actually are and do not have the metacognitive ability to realise they have done so.

Illusory inferiority is the flip side of that coin. It is tied up in levels of accurate internal and external self-awareness and is critical for establishing appropriately challenging goals.

Where are your limits?

We often set our goals based on our current abilities and past experiences without fully realising our true potential. If you think about it, how can anyone set a goal independently without precisely assessing their potential?

But the potential is much harder to assess. When you base your goals on past experiences, there will always be limits to what you aim for. A plan stretching your or your team’s capabilities beyond experience is much harder to imagine.

The concept of potential is deeply rooted in psychology. One of the fundamental theories in this area is the ‘Self-efficacy theory’ proposed by psychologist Albert Bandura.

Self-efficacy refers to an individual’s belief in their capacity to execute behaviours necessary to produce specific performance attainments. I didn’t know what running 42-50km felt like, but I believed I could do it.

This belief reflects confidence in the ability to exert control over one’s own motivation, behaviour, and social environment. Perhaps this also loops back to the flipping of a mental switch, as I mentioned earlier.

It is also worth noting the arbitrary nature of the goals we establish and the structures we put up to scaffold our progress. Why a neatly rounded 50km? Why not 57km or 61km? I notice this ceiling, which paradoxically acts as something to aim for and a limit to what we can achieve.

Smile on perils past

Let’s finish this exploration by surfacing a process I have shared in this newsletter. Reframing or shifting our perspective is a powerful learning experience, which is true of endurance running as it is in educational growth.

And it starts by shifting the sands of time, from deep in the past to your present, and then onto an imagined future state.

Retrospective ➜ look back

“Look back and smile on perils past”. Every step you have taken has led you to this moment. Your experiences do not need to be disconnected. Accept and build on the lessons from the past.

Every challenging run was an opportunity for me to learn and grow stronger for what’s to come.

Introspective ➜ look inward

Look inside yourself to notice what is happening with your disposition. Increase your self-awareness of the ups, downs and spirals.

When I was running I experienced periods of doubt and discomfort, but I stuck at it, and achieved my goal.

Prospective ➜ look ahead

We scan the horizon and look ahead. Transform your insights into options and actions. Explore the limits of what is achievable.

I like to think of the question, “who is the person running at their potential?”

~

The journey I am on, running beyond what I previously believed to be my limits, has been a vivid metaphor for goal setting and personal growth.

The key takeaway is that our perceived limitations are often self-imposed, stemming from underestimating our true capabilities.

I encourage you to challenge these self-inflicted boundaries, whether in a professional environment, personal growth, or while lacing up your running shoes.

~

Let’s finish with some next steps, talking points and curated readings to take you further.

⏭🎯 Your Next Steps
Commit to action and turn words into works

Identify any invisible “glass ceilings” or assumptions that are limiting the goals you set for yourself and your team to what you have achieved before.

Get an outside perspective from a mentor or coach to realistically assess your full potential and capabilities beyond your own biased self-perception.

Set stretch goals for yourself that purposefully go beyond your comfort zone, aiming higher than before to tap into your latent potential.

🗣💬 Your Talking Points
Lead a team dialogue with these provocations

⬩ What biases or thought patterns might be causing us to unconsciously underachieve as individuals and as an organisation?

⬩ How can we foster a culture that motivates ambition, embraces challenges, and believes in each person’s potential?

⬩ What examples or role models inspire you to envision your own capabilities beyond what you previously thought possible?

🕳🐇 Down the Rabbit Hole
Still curious? Explore some further readings from my archive

The Shape of the Lens (Tom Barrett) Inspired by an optics metaphor used in ethnography, this mental model explores our perception and understanding of behaviour. Use it as part of your team’s developmental dialogue and process.

Forget About Setting Goals. Focus on This Instead (James Clear) I set goals for the grades I wanted to get in school, for the weights I wanted to lift in the gym, for the profits I wanted to earn in business. I succeeded at a few, but I failed at a lot of them. Eventually, I began to realize that my results had very little to do with the goals I set and nearly everything to do with the systems I followed.

10 Shifts in Perspective To Unlock Insight and Embrace Change (Tom Barrett) The skills, dispositions and routines of shifting perspectives are potent catalysts to better thinking and dialogue. Here is a selection of perspectives to explore.

Thanks for reading. Drop me a note with any Kind, Specific and Helpful feedback about this issue. I always enjoy hearing from readers.

~ Tom Barrett

humAIn

A learning community for educators to connect and explore our humanity in the age of artificial intelligence.

Limited places for Australian educators.


*Applications close October 1, 2023

Reflective Practice Workbook

With this workbook as part of a healthy habit of thinking:

  • Reduce rumination and overwhelm.
  • Clarify where you have the most significant impact.
  • Gain a better understanding of your strengths.

Debrief and reflect after lessons, events or workshops, as part of project reviews and cycles of mentoring reflection.

+ BONUS resource: 100 Journal Prompts – this bonus PDF bundles with the workbook.

Workbook Includes: thinking routine templates, step by step guide, questions prompts, action planning, reflective habit tracker, further readings and resource links.

Promptcraft

.: AI for a better learning ecosystem :.

Are you fascinated by the world of AI for education
but finding it impossible to keep up to date with everything?

I also curate a newsletter called Promptcraft, everything you need to know about artificial intelligence for education.

Start your week ahead of the curve with the latest educational thinking and news in artificial intelligence.

If this sounds like something you’d enjoy, click the link below and enter your email address.

Support this newsletter

Donate by leaving a tip

Encourage a colleague to subscribe​​​​​

Tweet about this issue

The Bunurong people of the Kulin Nation are the Traditional Custodians of the land on which I write and create. I recognise their continuing connection and stewardship of lands, waters, communities and learning. I pay my respects to Indigenous Elders past, present and those who are emerging. Sovereignty has never been ceded. It always was and always will be Aboriginal land.

Unsubscribe | Update your profile | Mt Eliza, Melbourne, VIC 3930

.: Promptcraft 26 .: UNESCO publishes the first global guidance on how to use generative AI tools

Hello Reader,

Welcome to Promptcraft, your weekly newsletter on artificial intelligence for education. Every Monday, I curate the latest news, tools and resources so you can consider how AI changes how we teach and learn.

In this issue:

  • Anthropic introduces Claude Pro
  • UAE launches Arabic large language model
  • UNESCO publishes the first global guidance on how to use generative AI tools

Let’s get started!

.: Tom

Latest News

.: AI Updates & Developments

.: AI technology behind ChatGPT was built in Iowa — with a lot of water ➜ Microsoft and OpenAI have built a powerful supercomputer in West Des Moines, Iowa, to train their generative AI models. Microsoft’s global water consumption increased by 34% from 2021 to 2022, mostly due to its AI work. Google also reported a 20% growth in water use in the same period.

.: Google: Political adverts must disclose use of AI ➜ Google has announced that political ads on its platforms will soon have to disclose if they have used artificial intelligence to create or modify images, video or audio content. This new rule is a response to the increasing prevalence of tools that generate synthetic content, also known as deepfakes. The policy will come into effect in November 2023.

.: UAE launches Arabic large language model in Gulf push into generative AI ➜ Jais is an open-source Arabic language model developed by G42, in collaboration with Mohamed bin Zayed University of Artificial Intelligence and Cerebras Systems. It is considered the world’s highest-quality Arabic AI software, powering generative AI applications for Arabic-speaking users.

g7njDdJr5H1FwnoNCanJuD

.: Ideogram launches AI image generator with impressive typography ➜ Ideogram is a startup founded by former Google Brain researchers that offers a web app for generating images with different styles, including typography.

.: UNESCO: Governments must quickly regulate Generative AI in schools ➜ UNESCO publishes the first global guidance on how to use generative AI tools, such as ChatGPT, in education and research in a human-centred and ethical way.

.: Anthropic Introducing Claude Pro ➜ Anthropic have announced Claude Pro: A paid plan for ​Claude.ai​ that offers 5x more usage of the latest AI model, Claude 2, for $20 (US) or £18 (UK) per month. Subscribers get early access to new features and can use Claude 2 for various tasks, such as summarizing research papers, querying contracts, and coding projects.

.: New Open-Source ‘Falcon’ AI Language Model Overtakes Meta and Google ➜ Falcon 180B is a new open-source large language model (LLM) with 180 billion parameters trained on 3.5 trillion tokens. It surpasses previous open-source LLMs in both scale and performance, and reaches near parity with commercial models like Google’s PaLM-2. Falcon 180B is available on Hugging Face Hub, an online platform for sharing and collaborating on AI models.

.: AI reading coach startup Ello raises $15M to bolster child literacy ➜ Ello is a subscription-based service that delivers five books per month for children from kindergarten to Grade 3. The service uses AI and child speech recognition technology to listen to children read aloud and provide feedback and guidance. Ello has raised $15 million in Series A financing to further develop its product and expand access to consumers.

Reflection

.: Why this news matters for education

The past week saw UNESCO bustling with activity as it hosted the inaugural edition of “Digital Learning Week” in Paris, focusing on Generative AI.

Along with publishing the first global guidance on how to use generative AI tools – much needed – there were announcements about competency frameworks for teachers and students.

There appears to be a significant gap in the realm of education-specific structures, frameworks, and regulations. While I am a staunch advocate for innovation, it is equally crucial to acknowledge that new ideas need the backing of reliable, existing structures to flourish.

I find the interdependence of new and existing practices fascinating. In this case, we need stabilisers around the new ideas emerging from artificial intelligence tools.

I wonder how these new frameworks and guidelines can help articulate and propagate the potential of these emerging ideas. These frameworks could make the paradigm shift more digestible, facilitating greater understanding.

The fact that such frameworks are being released by an authoritative body like UNESCO lends significant credibility to these burgeoning practices, encouraging individuals yet to engage fully.

The proposed competency framework employs the language of Bloom’s Taxonomy for progression. This is not my preferred approach, as it emphasises a hierarchical and sequential learning arrangement. Learning can often be messy; this is particularly true for complex AI tools, which can provide high utility even without a foundational understanding.

As I delve deeper into the wealth of information from UNESCO’s Digital Learning Week, I will highlight more critical points in the upcoming weeks.

The bottom line is that introducing human-centred frameworks for communication, regulation, competency, and care is a positive step forward. It provides the necessary foundation upon which innovation can grow and prosper.

.:

~ Tom

Prompts

.: Refine your promptcraft

Today we have a little prompt to use at the beginning of your instructions. It is another example of a prompt that triggers the concept of Chain-of-thought (CoT) reasoning, which has been shown to deliver better responses.

Last week I highlighted how Anthropic illustrated this by including <thinking> tags to encourage planning and intermediate reasoning.

Today’s prompt to try is from a paper from Google Deepmind and highlighted by Ethan Mollick.

Add this to the start of your instructions:

Take a deep breath and work on this problem step-by-step.

I know this sounds odd but when used this was shown to produce much better results. The other prompts in the top 4 were:

  • Break this down.
  • A little bit of arithmetic and a logical approach will help us quickly arrive at the solution to this problem.
  • Let’s combine our numerical command and clear thinking to quickly and accurately decipher the answer.

Any of these will be better than not using them. Remember there is no breathing or thinking involved when a LLM is completing the task for you.

Weird and curious how it responds to such a human command, perhaps something to do with the nature of the training data.

.:

Remember to make this your own, tinker and evaluate the completions.

Learning

.: Boost your AI Literacy

.: Welcome to LLM University! (cohere.com)

“The course covers LLMs starting from the basics, all the way to building and using text representation and text generation models. Its theoretical portion is explained clearly and with analogies and examples rather than formulas, and its practical portion contains lots of useful code examples that will help you solidify your knowledge. “

.: Guidance for generative AI in education and research – UNESCO Digital Library

“The publication offers concrete recommendations for policy-makers and education institutions on how the uses of GenAI tools can be designed to protect human agency and genuinely benefit students, teachers and researchers.”

.: Digital Learning Week | UNESCO

Resources, replays and frameworks from UNESCO’s Digital Learning Week which looked closely at public digital learning platforms and generative AI, examining how both can be steered to reinforce and enrich humanistic education.

*There is a lot here including keynote talks from the likes of Yoshua Bengio and Yann LeCun and new competency frameworks for Teachers and Students, so we will return to more of these highlights in the weeks ahead.

Ethics

.: Provocations for Balance

While ChatGPT reached 100 million monthly active users in January 2023, only one country has released regulation on generative AI in July 2023.

~ Inspired by this week’s developments.

.:

That’s all for this week; I hope you enjoyed this issue of Promptcraft. I would love some kind, specific and helpful feedback.

If you have any questions, comments, stories to share or suggestions for future topics, please reply to this email or contact me at tom@dialogiclearning.com

The more we invest in our understanding of AI, the more powerful and effective our educational systems become. Thanks for being part of our growing community!

Please pay it forward by sharing the Promptcraft signup page with your networks or colleagues.

.: Tom Barrett

/Creator /Coach /Consultant

🔦 Reverse the Curse: 3 Biases in the Spotlight

Dialogic #332

Leadership, learning, innovation

Your Snapshot
A summary of the key insights from this issue

⬩ The curse of knowledge bias can undermine communication when we mistakenly assume others share our level of expertise. Avoiding jargon and checking for understanding helps reverse this.

⬩ The streetlight effect limits innovation, as we tend to only search for solutions in familiar places rather than exploring broadly.

⬩ Confirmation bias leads us to favour information aligning with pre-existing views, while dismissing contradicting evidence. Seeking divergent perspectives counters this.

Reverse the Curse: 3 Biases in the Spotlight

In this week’s issue, we cycle back through some of our latticework’s most important mental models: cognitive biases.

A concise way of thinking about the three cognitive biases we are exploring today is:

  • What I know
  • Where I look
  • How I listen

What I Know – The Curse of Knowledge

The curse of knowledge is a cognitive bias that occurs when an individual assumes that an audience has the background knowledge to understand. This bias is also called the curse of expertise.

This can lead to ineffective or confusing communication, especially when the communicator and the audience have a gap in knowledge or experience.

For example, in a classroom setting, teachers may have difficulty if they cannot put themselves in the student’s position and what it means to have an emerging understanding of a topic.
Or when a school leader tries to explain a new policy or initiative to the staff but uses jargon or acronyms that are unfamiliar to them or skims over elements that are unclear but essential.

The school leader may assume that the staff knows what she is talking about, but in reality, they may be confused or unclear about the purpose or implications of the change. This can result in frustration, resistance, or misunderstanding among the staff and undermine the effectiveness of the leadership.

Let’s take a look at some strategies to Reverse the Curse!

  1. Check for understanding and feedback from your listeners.
  2. For higher-stakes interactions, pay close attention to the language you use.
  3. Challenge assumptions and seek clarification. As a listener, step up and clarify.
  4. Avoid jargon and technical language that may confuse or alienate your audience.
  5. Use analogies and metaphors to make your message more relatable and memorable.
  6. Learn more about your audience and the context they face. Tune into their experience.
  7. Anticipate questions, concerns, and interests, and address them in your communication.
  8. Ask someone less knowledgeable or familiar with your topic to review it and give feedback.
  9. Use concrete language, examples, and stories to make your message more accessible and engaging.

Where I look – The Streetlight Effect

The streetlight effect can explain a block to new ideas and innovative solutions. You might have heard of this observational bias, illustrated in the story of the drunk looking for his keys:

A policeman sees a drunk man searching for something under a streetlight and asks what the drunk has lost. He says he lost his keys, and they both look under the streetlight together. After a few minutes, the policeman asks if he is sure he lost them here, and the drunk replies no and that he lost them in the park. The policeman asks why he is searching here, and the drunk replies, “This is where the light is”.

Sometimes, people look for the next breakthrough idea in the most accessible place. ​They filter for ideas​ closely related to their work or too similar to their context. That search is doomed to mediocrity. At best, it was a marginal alteration, not the breakthrough they hoped for.

It may be easier to look at what the school down the road is doing, but that limits what is possible.

The streetlight effect is ​a helpful bias to reflect on​ when we develop potential solutions.

How I listen – Confirmation Bias

Confirmation bias is the tendency to search for, interpret, favour, and recall information in a way that confirms one’s preexisting beliefs or hypotheses. It can make it difficult to consider alternatives or dissenting perspectives.

This cognitive bias is highly relevant to school leadership for a few reasons:

  • Leaders may fall victim to confirmation bias in seeking data, input, and solutions that validate their views rather than considering a more comprehensive range of sources.
  • Confirmation bias can cause leaders to favour options that align with their assumptions and discount evidence to the contrary.
  • Leaders have to balance confirmation bias with authentic listening and learning. Being open to having one’s views challenged takes humility and metacognition.
  • In communication, leaders must be aware of how confirmation bias may cause their message to be interpreted differently by stakeholders with differing perspectives.
  • Creating organisational culture change requires openness to ideas counter to established norms and practices. Confirmation bias works against this.

Being aware of confirmation bias and its relationship to the Curse of Knowledge and the Streetlight Effect can help leaders be more open-minded, creative, and effective in decision-making and communication.

Let’s finish with some next steps, talking points and curated readings to take you further.

⏭🎯 Your Next Steps
Commit to action and turn words into works

Ask trusted advisors to point out your biases and assumptions. Be willing to listen.

Have an open discussion with your team about biases and blindspots. Share examples.

When making an important decision, intentionally seek dissenting views and alternative options.

🗣💬 Your Talking Points
Lead a team dialogue with these provocations

⬩ What voices are missing in our team conversations? Whose realities and contexts get overlooked in our planning?

⬩ How often do we seek input from those with vastly different contexts and perspectives? Are we operating in an echo chamber?

⬩ When was the last time we looked for innovative solutions outside our normal sources? Are we stuck searching under the “streetlight” of the familiar?

🕳🐇 Down the Rabbit Hole
Still curious? Explore some further readings from my archive

Counter Wooden-Headedness and Break Your Echo Chambers – Tom Barrett (edte.ch) In my article from last year I explore how teams and organisations can benefit from good conflict and collective intelligence, which are fostered by diverse and divergent perspectives.

The Shape of the Lens – Tom Barrett (edte.ch) One of the most important questions we can ask to highlight the biases present in our work is: What is the shape of the lens? Use it as part of your team’s developmental dialogue and process.

Are You Solving the Right Problems? (hbr.org) “What business leaders struggle with is not solving problems but figuring out what the problems are. And creative solutions nearly always come from an alternative explanation for—or a reframing of—your problem.”

Thanks for reading. Drop me a note with any Kind, Specific and Helpful feedback about this issue. I always enjoy hearing from readers.

~ Tom Barrett

Promptcraft

.: AI for a better learning ecosystem :.

Are you fascinated by the world of AI for education
but finding it impossible to keep up to date with everything?

I also curate a newsletter called Promptcraft, everything you need to know about artificial intelligence for education.

Start your week ahead of the curve with the latest educational thinking and news in artificial intelligence.

If this sounds like something you’d enjoy, click the link below and enter your email address.

Support this newsletter

Donate by leaving a tip

Encourage a colleague to subscribe​​​​​

Tweet about this issue

The Bunurong people of the Kulin Nation are the Traditional Custodians of the land on which I write and create. I recognise their continuing connection and stewardship of lands, waters, communities and learning. I pay my respects to Indigenous Elders past, present and those who are emerging. Sovereignty has never been ceded. It always was and always will be Aboriginal land.

Unsubscribe | Update your profile | Mt Eliza, Melbourne, VIC 3930