RISK REPORT

.: Election disruption from AI poses the biggest global risk in 2024, Davos survey warns​

Summary ➜ The World Economic Forum’s Global Risks Report 2024 has highlighted AI-derived misinformation and disinformation as the most significant global risk over the next two years. This concern is especially pertinent as approximately half of the world’s adult population is set to vote in upcoming elections, where AI’s influence on large voter populations could significantly impact democratic processes.

Why this matters for education ➜ The ongoing debate over what’s real and what’s not in education, primarily focused on plagiarism, is a distracting sideshow. This narrow focus shifts attention away from the critical need to develop robust skills against the emerging risk of the blurred line between truth and falsehood. In the AI era, it’s vital for students to learn how to discern misinformation and critically assess digital content. This skill is not just an academic necessity but a global imperative, as AI’s influence spans across international borders, reshaping political and social landscapes. Addressing this challenge requires a broader, more globally aware educational approach.

.: Other News In Brief
​

​Midjourney V6 is here with in-image text and completely overhauled prompting

​New material found by AI could reduce lithium use in batteries

OpenAI’s GPT Store Already Filling Up With “AI Girlfriends“

​Quora raises $75m for its AI chatbot platform

Rabbit sells out two batches of 10,000 R1 pocket AI companions over two days​

​Google AI has better bedside manner than human doctors — and makes better diagnoses

.: Join the community waitlist
​

There’s a special community on the horizon for educators like you who want to explore the human side of artificial intelligence.

In February, we’re opening up the humAIn community – a space for forward-thinking educators to connect and learn together as we navigate the age of AI.

By joining, you’ll:

• Build connections with like-minded peers
• Attend exclusive webinars and virtual events
• Join lively discussions on AI’s emerging role in education
• Access member-only resources and Q&A forums

It’s a chance to be part of something meaningful – a space to share ideas, find inspiration, and focus on our shared humanity.

Get your name on the waitlist for information, so you don’t miss out on being part of this new learning community about AI for education.

Let’s shape the future of education, together.

What’s on my mind?

.: The Faces We Long to See

​

Imagine this: You’ve just returned from a trip, having navigated the familiar airport routine – security lines, scanners, the usual. But this time, something strikes you differently as you clear the final checkpoint and shunt your luggage towards the exit. It’s not just seen; it’s felt.

.: :.

There I was, fresh off a flight, and I couldn’t help but notice something striking. Do you know those facial recognition systems at passport control? Impressive, sure. Machines whirring, beeping, scanning documents, recognising faces. They’re fast, they’re efficient. It’s technology at its peak, streamlining what used to be a long, human-driven process. Impersonal but effective. That’s the scene on one side of the airport.

When I moved beyond the systems of digital precision, the atmosphere shifted. As I pushed my luggage, here, in the arrivals hall, the scene transforms. Teenagers huddle together, smartphones in hand, homemade signs aloft – a buzzing hive, eagerly awaiting a friend’s return. Over there, a tearful couple, lost in the embrace of their children, a reunion that’s been long in the making.

For a moment, the room scanned me, and I could feel the collective anticipation – the expectant gazes of hundreds, each pair of eyes telling a story of waiting, of longing. I noticed the anxious grip on bouquets, flowers bunched in hands trembling with anticipation. This is more than just an arrivals hall; it’s the culmination of countless stories, the end of long countdowns, and the final moments of anticipation unfolding before our eyes.

They’re looking for faces, yes, but not just any faces – they’re searching for that one face they’ve missed and long to see. Hearts are racing; eyes are searching, and then a moment of recognition. It’s joy, it’s relief, it’s love. All happening right there, in the most human way possible. This was facial recognition powered by affection and memory, not algorithms.

This contrast, it hit me hard. On one side, machines do what they’re programmed to, precisely recognising faces. But they’re missing something crucial, something they can’t replicate – the emotion, the history, the storied connection we read in a human face. That’s our thing, our human thing.

And amid all the noise and rush, there’s a reminder in this bustling, busy airport. It’s a reminder of what makes us human, something that technology, no matter how advanced, can’t touch. The human connection, that spark when you see a familiar face, the warmth of a smile – technology might mimic it, but it can never truly capture it. Throughout history, our ability to recognise faces has evolved far beyond mere survival – it’s become a cornerstone of emotional connection and social interaction.

As I left the airport into the chilled Melbourne air, the echoes of these emotional reunions lingered with me. In our digital world, moments like these remind us that no matter how advanced technology becomes, the human ability to connect still holds irreplaceable value.

“But do you remember where we parked?”

:. .:

​

~ Tom

Prompts

.: Refine your promptcraft

​

Develop Scenarios for Critical Thinking

Scenario building is a great way to quickly resource some critical thinking activities. For example here is a scenario generated from today’s prompt about the morality of virtual worlds:

You will see from the longer prompt below I am using the structure:

• Persona / Role
• Task / Steps
• Format / Tone
• Context / Constraints
• Examples / Model Answers (optional)

Here is an example prompt, which aims to develop some critical thinking scenarios, for you to try.

PROMPT

Please note the variables you can change are included in parentheses.

This year I aim to share good examples of prompts as well as sharing with you new promptcraft techniques.

Remember to mak​e this your own, try different language models and evaluate the completions.

Learning

.: Boost your AI Literacy
​

LOOK AHEAD ​
​.: After AI’s summer: What’s next for artificial intelligence?

Ideogram prompt by VentureBeat

AI STRATEGY ​
​.: The secret to making language models useful​

Here is a summary of the key insights from the article and why it is useful for your AI Literacy:

• The main idea is that language models alone are not enough to be truly useful or make good decisions. Language provides the words, but you need knowledge and understanding to apply those words wisely.
• Language models can recite words and phrases from their training data, but they lack true comprehension. They find statistical correlations but can’t determine causation.
• To make language models useful, you need to recreate the structure of human expertise – combining language with knowledge and understanding. This means knowledge graphs, causal models, etc.
• The process should start with identifying the most valuable human expertise in an organization, determining the risks of losing it, and seeing if it can be encoded for machines.
• Data is not the most critical element – expertise is. The goal is transferring human expertise into machine language so machines can inform or make decisions.
• Leaders need to focus on what expertise could give them a competitive advantage if encoded into machines, rather than just reacting to what others develop.

This article provides an important perspective – that language AI like ChatGPT is not magic on its own. To be useful you need to carefully encode human knowledge, not just data.

For me this relates to the way Roger Martin talks about the Knowledge Funnel and how we don’t often have system to encode human expertise.

This is especially true in schools and back in 2015 I was calling this a knowledge urgency problem. Where we underinvest in really capturing the methods, tactics and strategies of the best expertise we have before they move on.

It is interesting to think that training a language model might be a vessel for that expertise.

​

WHICH AI IS BEST? ​
​.: People Rank Open Source AI Output Better Than Commercial Models ​

A couple of elements of your AI literacy which I encourage you to pay attention to this year are:

• What is it like to use open-source LLMs?
• Explore the performance comparison across available models.

There is a wide variety of models available and when you think about AI you don’t just think of ChatGPT, you might think – OK which models might be best for this task.

Here’s a snippet from this article which refers to leaderboard results for performance from HuggingFace.

Ethics

.: Provocations for Balance
​

Here’s the full text from the critical thinking scenario prompt which I thought was worth sharing in full.

.:

Philosophical Dilemmas in Virtual Realities

Scenario

Probing Questions:

1. Moral Parallels: In what ways might actions in a virtual environment ethically parallel those in the real world? Consider the implications of harm or benefit in both settings.
2. Reality vs. Perception: How does our understanding of ‘reality’ affect our moral judgments? Discuss whether the perceived reality of VR changes ethical considerations.
3. Consequence-Based Ethics: Should the lack of real-world consequences in VR impact our moral decision-making? Explore this through the lens of different ethical theories.

.:

Read the rest in this ChatGPT chat here.

Which section was the most helpful?

(Click on your choice below)
​

​📰 Curated news and updates​
​💭 My reflections and commentary​
​📚 Learning resources + links​
​✏️ Promptcraft techniques​
​🤔 Ethics provocations

​
.: Tom Barrett