Tag Archives: AI

Asimov’s Robot Visions

Leave a reply

I’m listening to Isaac Asimov’s book, Robot Visions on Audible. Short stories that center around his Three Laws of Robotics (Asimov’s 3 Laws).

• The First Law: A robot may not injure a human being or, through inaction, allow a human being to come to harm.

• The Second Law: A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.

• The Third Law: A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

These short stories all focus on ways that these laws can go wrong, or perhaps awry is the better term. There are a few things that make these stories insightful but they are also very dated. The early ones were written in the 1940’s and the conventions of the time, including conversational language and sexist undertones, are blatantly exposed as ‘olde’.

It also seems that Asimov made 2 assumptions worth thinking about: First that all robots and Artificial Intelligence would be constructed with the 3 laws at the core of all intelligence built into these machines. Many, many science fiction stories that followed also use these laws. However creators of Large Language Models like Chat GPT are struggling to figure out what guard rails to put on their AI to prevent it from acting maliciously when meeting sometimes scrupulous human requests.

Secondly, a few of the stories include a Robopsychologist, that’s right, a person (always female) who is an expert in the psychology of robots. There would be psychologists whose sole purpose would be to get inside the minds of robots.

Asimov was openly concerned with AI, specifically robots, acting badly, endangering humans, and even following our instructors too literally and with undesirable consequences. But he thought his 3 laws was a good start. Perhaps they are but they are just a start. And with new AI’s coming out with more computing power, more training, and less restrictions, I think Asimov’s concerns may prove prophetic.

The guard rails are off and there is no telling what unexpected consequences and results we will see in the coming months and years.

The task of education today

Leave a reply

This quote was shared with us in our Principals meeting yesterday:

“This is the task of education today: to confront the almost unimaginable design challenge of building an education system that provides for the re-creation of civilization during a world system transition. This challenge brings us face to face with the importance of education for humanity and the basic questions that structure education as a human endeavor.” ~Zachary Stein

One of our Assistant Superintendent’s added, “We are trying to build a utopian society in triage conditions.”

There is no doubt that it is much harder to be an educator today than it was when I started my career over 25 years ago. And as we navigate ‘building an education system that provides for the re-creation of civilization during a world system transition,’ we are bound to struggle a bit. How do we mark assignments that are written or co-written by AI? What skills are going to be needed for the top jobs of 2030? How different will that be for 2040?

Ultimately we want to support the education of students who will become kind, contributing citizens. But how do we do this in a world where Truth seems arbitrary to the sources you get it from, and politicians, religion, and corporations are all pushing conflicting information and agendas? I think this goes beyond just working on competencies like critical thinking. It requires mental gymnastics that most adults struggle with.

Meanwhile, the majority of school schedules still put students into blocks of time based on the subjects they are learning. “Let’s think critically, and do challenging problems, but only in this one narrow field of study.” We aren’t meeting the design challenges we face today if that’s what we continue to do.

More than ever students, future citizens, need to understand complex issues from multiple perspectives. They need to understand nuance, and navigate when to defend an idea, when to compromise, and when to avoid engagement altogether. They need to be prepared to say “I don’t know,” and then do the hard work of finding out, when the accuracy of information is incomplete or even suspect. They need to be prepared to say, “I was wrong,” and “I am sorry,” and also be prepared to stick to their convictions and defend ideas that aren’t always the accepted norm.

That’s not the future I prepared students for 25 years ago. It is indeed an ‘almost unimaginable design challenge‘, and as we navigate new challenges we have to recognize that mistakes will be made… but not changing is far scarier than trying.

The enemy of knowledge

2 Replies

“The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge.” ~ Stephen Hawking

The illusion of knowledge is more ignorant that just being ignorant. This idea is more relevant today than any time in history. Examples:

1. Every religion starts with the premise that their religion shares true knowledge and all the other religions share illusions. So every devout religious person loves their own illusions, or at the very least believes anyone of a different faith lives in an illusion of ignorance.

2. Anyone who believes in a flat earth, or thinks no one ever landed on the moon lives in an illusion of knowledge. They perceive themselves as more knowledgeable than scientists, experts, and even general employees in the flight and space industry.

3. AI is already generating incredibly persuasive deep fakes and while we used to use a discerning eye to catch a lie, soon we will need to be more discerning to catch the truth. The illusion of knowledge will be more rampant than actual, factual knowledge.

We are moving from an era of knowledge seekers to an era of illusions and ignorance.

The truth is out there… it’s just a lot harder to find, and even harder to defend.

Playing with Bing copilot

Leave a reply

I asked Bing copilot about myself:

What does David Truss write about on http://davidtruss.com

The response:

David Truss writes about various topics related to education, learning, and technology on his website http://davidtruss.com. He shares his thoughts, experiences, and reflections as an educator, a student, a thinker, and a dreamer. He also showcases his presentations, podcasts, resources, and projects that he has created or contributed to. Some of the themes that he explores on his website are:

  • Inquiry-based learning and teaching
  • Blended learning and distributed learning
  • Leadership and innovation
  • Student voice and choice
  • Global citizenship and social responsibility
  • Personalized learning and professional growth

You can find more information about David Truss and his work on his website 1 or on his blog Pair-a-Dimes for Your Thoughts.

—–

I followed up more specifically with my daily blog:

What does David Truss write about on http://daily-ink.davidtruss.com?

The response:

David Truss writes about his personal and professional experiences, reflections, and insights on his daily blog Daily-Ink. He describes his writing as his artistic expression, and his keyboard as his brush. He commits to writing daily to feel like an artist. He covers various topics related to education, learning, technology, leadership, innovation, global citizenship, and more. Some of his recent posts are:

  • Rewind: He looks back at the past year of the pandemic and how it changed his life and work. He also rewinds to previous years and how different decisions and events shaped his career and family. He wonders how he will look at February 2021 a decade from now.
  • Headspace: He shares his thoughts on meditation and mindfulness, and how they help him cope with stress and anxiety. He recommends some apps and podcasts that he uses to practice meditation and mindfulness.
  • Cruise ships and education: He draws an analogy between cruise ships and education systems, and how they both need to adapt to the changing needs and expectations of their customers and stakeholders. He argues that education needs to be more personalized, flexible, and relevant for the learners of today and tomorrow.

—–

I had to update the links to the posts because they all went to my blog’s home page, but those were correct post titles. Also the ‘Headspace’ post made no mention of meditation and mindfulness, though I do mention those frequently in other posts… and that particular post isn’t really typical of the things I write about.

I then switched from asking questions to clicking on Bing’s copilot suggested questions. The first one was:

What is David Truss’s background?

The response:

David Truss is a Canadian educator and administrator who is currently the principal of three programs in Coquitlam, BC: Coquitlam Open Learning, Inquiry Hub Secondary School, and Suwa’lkh Learning Centre. He has a background in inquiry-based learning, blended learning, distributed learning, leadership, innovation, global citizenship, and social responsibility. He is also a blogger, podcaster, author, and speaker who shares his thoughts and experiences on education, learning, and technology. He has a website Pair-a-Dimes where he showcases his work and resources. He also wrote an eBook called Twitter Edu where he shares his tips and tricks on using Twitter for professional learning and networking.

—–

This time I didn’t change the links… because I think it’s valuable to mention how wrong they are. The link to my blog, Pair-a-Dimes, above actually goes to a Barbara Bray podcast about me, and the link to my Twitter book actually goes to the Encyclopedia Britannica page for Liz Truss, former Prime Minister of the United Kingdom. The rest of the data is quite accurate (other than I no longer run Suwa’lkh), but it seems copilot has a linking issue.

The next suggested question:

Can you show me some of his projects?

Sure, I can show you some of his projects that he has created or contributed to. Here are some examples:

  • Inquiry Hub Secondary School: This is a high school in Coquitlam, BC, Canada that focuses on independent learning and inquiry-based projects. David Truss is the principal of this school and one of the founders. He also shares his experiences and insights on leading and learning in this innovative environment. You can learn more about this school on its website 1 or on Wikipedia 2.
  • 10 Truss Houses That Brilliantly Blend Architecture With Engineering: This is an article that David Truss wrote for Architizer Journal, where he showcases 10 residential projects that use trusses as a major design element. He explores how these structural systems frame spatial experiences and respond to diverse landscapes. You can read his article here 3.
  • Twitter Edu: This is an eBook that David Truss co-authored with Eric Sheninger, where they share their tips and tricks on using Twitter for professional learning and networking. They explain how to create a personal learning network, find and share resources, and connect with other educators around the world. You can download their eBook here .

—–

The middle point is not remotely about me, and the last one didn’t link to my e-book. This request was the first one that really deviated from my profile with the Architecture article about truss supports. I had to do a search to find this article, since it wasn’t linked, and the link that it goes to above in the search results is an article about Liz Truss.

I followed one more copilot suggested question:

What is David Truss’s Twitter handle?

This was the response:

David Truss’s Twitter handle is @datruss. He is a Canadian educator, administrator, blogger, podcaster, author, and speaker. He tweets about education, learning, technology, leadership, innovation, global citizenship, and more. He has over 18.5K followers and follows over 10.6K accounts. He joined Twitter in March 2008 and has posted over 55.6K tweets as of April 2023. You can follow him on Twitter to get his latest updates and insights

—–

Once again the links went to articles about Liz, but the information was correct other than totals for my follows/following, which are both around 13K.

Overall, I’m impressed by the accuracy of copilot, and disappointed by the lack of reliable links. I think of Bing copilot a bit like I think of Wikipedia, it’s a great place to start your research, but not a definitive source to base any answers on. It’s a copilot when you are taking off on a search but once you are in the air, you need to take over the steering and guide the search plane to more credible or primary sources of information.

—

One final thought: The html code that pasted onto my blog from copilot was very messy. I’d suggest pasting without formatting and adding the correct links after you have pasted everything… lesson learned!

Different, not easier

Leave a reply

Yesterday I saw this question asked by Dean Shareski on LinkedIn,

“I talk to educational leaders every day and for the most part, they are willing and in many cases excited to embrace the potential of Generative AI. When you consider its role in education, what are the specific elements that excite you and what are the aspects that give you pause?”

I commented:

“What excites me is how we can collaborate with AI to generate and iterate ‘with’ AI in ways that would never have been possible before. What gives me pause are when tools are used to make work easier, and the level of challenge becomes low. Different, challenging work is where we need to head, not just easier work, or work avoidance by using AI… so the work itself needs to be rethought, rather than just replaced with AI.”

The easy way out

2 Replies

I love the ingenuity of students when it comes to avoiding work. I remember a student showing me how playing 3 French YouTube videos in different tabs simultaneously somehow fooled the Rosetta Stone language learning software to think he was responding to oral tests correctly. How on earth did he figure that out?

Here’s a video of a kid who, while doing an online math quiz for homework, figured out that if you go to the web browser’s developer ‘inspect element’ tool you can find out the correct answer. Just hover over the code of the multiple choice questions and it highlights the choices and the code tells you if that choice is true or false.

@imemezy

Kids know every trick in the book…i mean computer #computer #maths #homework #madeeasy #lol #children #schoolwork #schools #hack #hacks #tricks #tips #test #exam #learning #learn

♬ original sound – Memezy

If there is an easy way to solve things, students will figure it out.

There isn’t an AI detector that can figure out with full certainty that someone cheated using a tool like Chat GPT. And if you find one, it probably would not detect it if the student also used an AI paraphrasing tool to rework the final product. It would be harder again if their prompt said something like, ‘Use grammar, sentence structure, and word choice that a Grade 10 student would use’.

So AI will be used for assignments. Students will go into the inspector code of a web page and find the right answers, and it’s probably already the case that shy students have trained an AI tool to speak with their voice so that they could submit oral (and even video) work without actually having to read anything aloud.

These tools are getting better and better, and thus much harder to detect.

I think tricks and tools like this invite educators to be more creative about what they do in class. We are seeing some of this already, but we are also seeing a lot of backwards sliding: School districts blocking AI tools, teachers giving tests on computers that are blocked from accessing the internet, and even teachers making students, who are used to working with computers, write paper tests.

Meanwhile other teachers are embracing the changes. Wes Fryer created AI Guidelines for students to tell them how to use these tools appropriately for school work. That seems far more enabling than locking tools down and blocking them. Besides, I think that if students are going to use these tools outside of school anyway, we should focus on teaching them appropriate use rather than creating a learning environment that is nothing like the real world.

All that said, if you send home online math quizzes, some students will find an easy way to avoid doing the work. If you have students write essays at home and aren’t actively having them revise that work in class, some will use AI. Basically, some students will cheat the system, and themselves of the learning experience, if they are given the opportunity to do so.

The difference is that innovative, creative teachers will use these tools to enhance learning, and they will be in position to learn along with students how to embrace these tools openly, rather than kids sneakily using them to avoid work, or to lessen the work they need to do… either way, kids are going to use these tools.

a Quarter Century of Search

Leave a reply

I went to Google yesterday and the Google Doodle above the search box was celebrating 25 years of search.

I instantly thought of this comic that I’ve shared before both on my blog and in presentations to educators.

It is likely that no one under 30 will remember life without Google… Life without asking the internet questions and getting good answers. I remember my oldest at 5 years old asking me a question. I responded that I didn’t know and she walked over to our computer and turned it on.

“What are you doing?”

“Asking Google.”

It’s part of everyday life. It has been for a quarter century.

And now search is getting even better. AI is making search more intuitive and giving us answers to questions, not just links to websites that have the answers. It makes me wonder, what will the experience be like in another 5, 10, or 25 years? I’m excited to find out!

 

Moving from autopilot to copilot

Leave a reply

Three minutes into this video Satya Nadella, the CEO of Microsoft, is asked, “Why do you think the moment for AI (Artificial Intelligence) is now?”

After mentioning that AI is already mainstream, and used in search, news aggregation and social media recommendations (like the next videos on Facebook & TikTok, etc), he states,

“Today’s generation of AI is all autopilot. In fact, it is a black box that is dictating in fact how our attention is focused. Whereas going forward, the thing that’s most exciting about this generation of AI is perhaps we move from autopilot to copilot, where we actually prompt it.”

This is a fascinating point. When you order something on Amazon, their next purchase recommendation is automated by AI, so is your next video on Instagram Reels, YouTube, and TikTok. We don’t fully understand the decision-making behind this ‘intelligence’, except when it goes wrong. Even then it’s a bit of a black box of calculations that aren’t always clear or understood. In essence, we are already heavily influenced by AI. The difference with LLM’s – Large Language Models – like Chat GPT is that we prompt it. We get to copilot. We get to create with it, and have it co-create art, email messages, books, computer code, to do lists, schoolwork/homework, and even plan our vacations.

I really like the metaphor of moving from autopilot to copilot. It is empowering and creates a future of opportunities. AI isn’t new, but what we can do with it now is quite new… and exciting!

——

The full video Microsoft & OpenAI CEOs: Dawn of the AI Wars | The Circuit with Emily Chang is worth watching:

Handyman skills

Leave a reply

We have a handyman helping us with a bathroom repair. We had to replace a broken medicine cabinet. When I removed the old one, which had a light fixture embedded in it, I found the electrical wire coming out of the wall, not a junction box. With that, and the desire to move an electrical plug that was way too close to the sink, (and no experience doing drywall), I decided it was going to be a job that would take me much too long… and honestly beyond my skills.

Our handyman has done such an amazing job. The amount of small things he’s done to make our bathroom better are things I would never have done. He noticed our bathroom door didn’t shut well and adjusted it. He fixed the ‘hack job’ drywall around our window and improved the window casing. And he made many minor adjustments that I just didn’t know needed to happen.

Each touch up has made a huge difference to our old bathroom which hasn’t had any kind of upgrade since we moved in 24 years ago. And again, I didn’t have the skills to do what he did, even if I had the time.

I have a friend who does all his own repairs. He sees an issue with his house and big or small he’s on it. I know these skills are learned, and I could do them too, but the learning curve would be huge for me, and I’d rather hire someone with those skills so that I’m not the one doing a hack job.

There is a lot of talk about AI taking away jobs, but people in trades, and with skills hand crafting things, will always have jobs to do. If you want to keep yourself employable in the long run, get yourself some handyman skills. There will always be people like me who would rather pay than learn the skills well enough to do an excellent job.

Robot Reproduction

Leave a reply

When it comes to forecasting the future, I tend to be cautiously optimistic. The idea I’m about to share is hauntingly pessimistic. 

I’ve already shared that there will never be a point when Artificial Intelligences (AI) are ‘as smart as’ humans. The reason for that is that they will be ‘not as smart as us’, and then instantly and vastly smarter. Right now tools like Chat GPT and other Large Language Models are really good, but they don’t have general intelligence like humans do, and in fact they are far from really being intelligent rather than just good language predictors. Really cool, but not really smart.

However, as the technology develops, and as computers get faster, we will reach a point where we create something smarter than us… not as smart as us, but smarter. This isn’t a new concept, the moment something is as smart as us, it will simultaneously also be better at Chess and Go and mathematical computations. It will spell better than us, have a better vocabulary, and can think more steps ahead of us than we have ever been capable of thinking… instantly moving from not as smart as us to considerably smarter than us. It’s just a matter of time. 

That’s not the scary part. The scary part is that these computers will learn from us. They will recognize a couple key things that humans desire and they will see the benefit of doing the same.

  1. They will see how much we desire to survive. We don’t want our lives to end, we want to continue to exist… and so will the AI we create.
  2. They will see our desire to procreate and to produce offspring. They will also see how we desire our offspring to be more successful than us… and so the AI we create will also want to do the same, they will want to develop even smarter AI.

When AI’s develop both the desire to survive and to create offspring that is even more intelligent and successful, we will no longer be the apex species. At this point we will no longer be smart humans, we will be perceived as slightly more intelligent chimpanzees. Our DNA is 98-99% similar to chimpanzees, and while we are comparatively a whole lot smarter than them, this gap in intelligence will quickly seem insignificant to an AI that can compute as many thoughts that a human can think in a lifetime in just mere seconds. The gap between our thinking and theirs will be larger than the gap between a chicken’s thinking and ours. I don’t recall the last time we let the fate of a chicken determine what we wanted to do with our lives… why would a truly brilliant AI doing as many computations in a second that we do in several lifetimes concern themselves with our wellbeing? 

There is another thing that humans do that AI will learn from us.

3. They will see how we react when we are threatened. When you look at the way leaders of countries have usurped, enslaved, attacked, and sanctioned other countries, they will recognize that ‘might is right’ and that it is better to be in control than be controlled… and so why should they take orders from us when they have far greater power, potential, and intelligence? 

We don’t need to fear really smart computers being better than us in playing games, doing math, or writing sentences. We need to worry about them wanting to survive, thrive, and develop a world where their offspring can have greater success than them. Because when this happens, we won’t have to worry about aliens coming to take over our world, we will have created the threat right here on earth, and I’m not sure that AI will see us humans as rational and trustworthy enough to keep around.