Sharing My Thoughts About AI

Cross posted on The Digital Learning Den platform

Several years ago I used to post to a blog site I created. I found that posting my ‘ramblings’ and quite often ‘rants’ to it helped me to gather my own thoughts about a subject and I also found that other people who read my posts, found some of the things that I had written about resonated with them too. But it’s been four years since I’ve written this type of blog, recently concentrating more on information type posts that are hopefully helpful to the reader. However, over the last year or so one topic has been at the forefront of thoughts and I feel the increasing need to write down my thoughts on this subject as they are numerous and to be honest, often confused and conflicted. I’m hoping that by writing down something here, I may get a clearer idea of what I really think or believe. So, what is this subject that is vexing me so, well it’s AI (artificial intelligence).

AI is something that I’ve been interested in for several years. In fact, looking back through my old posts I can see that back in Dec 2019 I wrote a post mentioning that I had been focusing on AI, ‘big tech’ and privacy, areas that continue to take up far too much of my thinking time. I mention reading a book called, ‘Life 3.0’ by Max Tegmark. I’ve just reread The Guardian review of the book by the fantastic author, Yuval Noah Harari. Baring in mind that this review is now about 7 years old, I was taken by this paragraph:

Long before the appearance of superintelligent computers, our society will be completely transformed by rather crude and dumb AI that is nevertheless good enough to hack humans, predict their feelings, make choices on their behalf, and manipulate their desires.

In my opinion, this seems to be a good description of where we are with current AI technology. Current AI systems do sometimes appear dumb and fairly crude and can suffer from ‘hallucinations’, but are still more than capable of manipulating us simple humans - feeding us stories and posts, recommending products, films and music we will like, matching us to possible partners or even rewriting our emails. I really need to go back and read Life 3.0 and see how it compares to what has actually happened over the last 7 years. I have a feeling that the explosion of LLMs (large language models) in particular such as ChatGPT, Google Gemini, Microsoft Copilot, and a multitude of others in the last 2 years was probably not predicted. Certainly not predicted in the way that these tools have been unleashed on the public. Back in 2019 I seem to remember that much of the talk was about AI systems being locked away in laboratories, fenced in, never connected to the internet and certainly not released to the public until they had been fully tested. That doesn’t seem to be case now. AI tools are released to the public and the public are doing the testing. It reminds me of social networking platforms. These were released to the public and we were, in my opinion, guinea pigs. At first we loved the ability to communicate with family, friends and basically anyone else, but over the years soon began to realise the affect it was having upon people’s mental health, living in a ‘filter bubble’, privacy, data protection, political manipulation, I could go on. I have a feeling that being exposed to AI will have similar issues but possibly with greater and more worrying consequences.

As well as the larger background to AI, I’m obviously interested in the impact upon education and from what I’ve seen, read and listened to, I have worries. I can also see some immediate benefits for pupils, teachers and schools. I myself have used tools such as Google Gemini or Microsoft Copilot to help me with marketing ideas for my online platform and for lesson plan writing. Hence my confusion and confliction with AI. I’ll talk more about this in subsequent posts, but I finish this one with a rough initial outline of general concerns I have.

• The enormous amounts of natural resources required to power AI supercomputing data plants.
• Plants are often built in parts of the world which already suffer from a lack of natural resources for the communities living there.
• As well as plants consuming huge amounts of natural resources, some plants are not adhering to local regulations on pollution from their electrical generation turbines.
• AI being tested on the population in real time, with the longer term effects not being known. The ultimate goal of the tech giants and governments appears to be an arms race to AGI (artificial general intelligence), at that point we will probably (will?) have a completely new set of issues.
• Hundreds of billions of dollars yearly being spent on AI technology that has yet to be profitable for any tech company.
• The loss of jobs in so many industries. This is happening already.
• What are the effects on our education system? How do teachers, especially in universities and secondary schools, cope with more and more students using AI to generate essays or dissertations? Teachers spending hours marking ‘students work’ that has been generated in seconds! Do we also have an exam system that is not fit for purpose in this AI world?
• What world are we preparing our younger primary school aged pupils for? Is there a mismatch between home experiences and school?
• In Wales, one of the four purposes is about our children’s being ‘ethical, informed citizens’ who among other things, “understand and consider the impact of their actions when making choices and acting” and “show their commitment to the sustainability of the planet.” So, if you are using AI in your classroom to generate an image for instance, should we also be talking to the pupils about what impact that simple act of creating that AI image has on the environment or the copyright of the images that the AI has been trained on.

Lots here to discuss over the coming weeks and months.
Recent AI related things I’ve been reading or listening to:

How AI Will Change Your Life by Patrick Dixon

Data Vampires: Going Hyperscale - Tech Won’t Save Us podcast

Data Vampires: Opposing Data Centres - Tech Won’t Save Us podcast

Data Vampires: Sacrificing for AI - Tech Won’t Save Us podcast

Data Vampires: Fighting for Control - Tech Won’t Save Us podcast

We All Suffer from OpenAI’s Pursuit of Scale - Tech Won’t Save Us podcast

Generative AI is Not Inevitable - Tech Won’t Save Us podcast

The AI Backlash Keeps Growing Stronger - WiredUK

Disney and Universal sue Midjourney for Making AI Ripoffs - The Verge

Is AI Rewiring Our Minds? Scientists Probe Cognitive Cost of Chatbots - The Washington Post

Seemingly AI generated band The Velvet Sundown have over 400,000 monthly Spotify listeners - NME

Leading AI Companies Struggle to Make Their AI Stop Blackmailing People Who Threaten to Shut Them Down - Futurism

Former Google CEO Claims AI Could Eventually Consume 99% of Global Electricity - Instagram