Tag Archives: technology

Speed and collar colour

Leave a reply

Two things are happening simultaneously.

First, the advancements we see in AI are moving at an exponential rate. Humans don’t really understand how to look at exponential growth because everything in front of them moves faster than what they’ve already experienced.

How many years did it take from the time light bulbs were invented until they were in most houses? I don’t know, but it took a long while. Chat GPT was used by over 100 million people in less than 2 months. And the ability of tools like this are increasing in ability exponentially as well. It’s like we’ve gone warp speed from tungsten and incandescent lights to LED’s in a matter of months rather than years and years.

The other thing happening right now is that for the first time at scale, it’s white collar, not blue collar jobs that are being threatened. Accountants, writers, analysts, coders, are all looking over their shoulder wondering when AI will make most of their jobs redundant. Meanwhile, we are many years away from a robot trying to figure out and repair a bathroom or ceiling leak. Sure, there will be some new tools to help, but I don’t think a plumbing home repair person is something AI is threatening to replace any time soon.

These two things happening so quickly are going to change the future value in careers. Whole sectors will be reinvented. New sectors will emerge. But where does that leave the 20-year accountant in a large firm that finds it can cut staffing by 2/3rds? What careers are not going to be worth going to university for 4+ years for? The safest jobs right now are the trades, and while they too will be challenged as we get AI into autonomous humanoid robots, the immediate threat seen in white collar jobs are not the same for blue collar professions (as opposed to blue collar factory workers, who are equally threatened by exponential changes).

These changes are single-digit years as opposed to decades away… and I’m not sure we are ready to handle the speed at which these changes are coming?

Don’t believe the hype

Leave a reply

The open source DeepSeek AI model has been built my the Chinese for a few million dollars, and it seems this model works better than the multi-billion dollar paid version of Chat GPT (at about 1/20th the operating cost). If you watch the news hype, it’s all about how Nvidia and other tech companies have taken a huge financial hit as investors realize that they don’t ‘need’ the massive computing power they thought they did. However, to put this ‘massive hit’ into perspective let’s look at the biggest stock market loser yesterday, Nvidia.

Nvidia has lost 13.5% in the last month, most of which was lost yesterday.

However, if you zoom out and look at Nvidia’s stock price for the last year, they are still up 89.58%!

That’s right, this ‘devastating loss’ is actually a blip when you consider how much the stock has gone up in the last year, even when you include yesterday’s price ‘dive’. If you bought $1,000 of Nvidia stock a year ago, that stock would be worth $1,895.80 today.

Beyond that, the hype is that Nvidia won’t get the big orders they thought they would get, if an open source LLM (Large Language Model) is going to make efficient, affordable access to very intelligent AI, without the need for excessive computing power. But this market is so new, and there is so much growth potential. The cost of the technology is going down and luckily for Nvidia, they produce such superior chips that even if there is a slow down in demand, the demand will still be higher than their supply will allow.

I’m excited to try DeepSeek (I’ve tried signing up a few times but can’t get access yet). I’m excited that an open source model is doing so well, and want to see how it performs. I believe the hype that this model is both very good and affordable. But I don’t believe the hype that this is some sort of game-changing wake up call for the industry.

We are still moving towards AGI, Artificial General Intelligence, and ASI – Super Intelligence. Computing power will still be in high demand. Every robot being built now, and for decades to come, will need high powered chips to operate. DeepSeek has provided an opportunity for a small market correction but it’s not an innovation that will upturn the industry. This ‘devastating’ stock price losses the media is talking about is going to be an almost unnoticeable blip in stock prices when you look at the tech stock prices a year or 5 years from now.

It is easy to get lost in the hype, but zoom out and there will be hundreds of both little and big innovations that will cause fluctuations in stock market prices. This isn’t some major market correction. It’s not the downfall of companies like Nvidia and Open AI. Instead, it’s a blip in a fast-moving field that will see some amazing, and exciting, technological advances in the years to come… and that’s not just hype.

The false AGI threshold

Leave a reply

I think a very conservative prediction would be that we will see Artificial General Intelligence (AGI) in the next 5 years. In other words, there will be AI agents, working with recursive self-improvement, that will learn how to do new tasks outside the realm of its training, faster than a human could. But when this actually happens will be open for debate.

The reason for this is that there isn’t going to be some magical threshold that AGI suddenly passes, but rather a continuous moving of the bar that defines what AGI is. There will be a working definition of AGI that puts up an artificial threshold, then an AGI model will achieve that definition and experts will admit that this model surpasses that definition, but will still think the model lacks some sufficient skills or expected outputs to truly call it AGI.

And so over the next 5 years or so we will have far more sophisticated AI models, all doing things that today we would define as AGI but will not meet the newest definition. The thing is that these moving goal posts will not be adjusted incrementally but rather exponentially. What that means is that the newer definition of AGI is going to include significantly greater output expectations. Then looking back, we will be hard pressed to say “this model’ was the one or ‘this day’ was the day that we achieved AGI.

Sometime soon there will be an AI model put out into the world that will build its own AI agent that starts a billion dollar company without the aid of a human… but that might happen even before consensus that AGI has been achieved. There will be an AI agent that costs lives or endangers lives with its decisions made in the real world, but that too might happen before consensus that AGI has been achieved.

Because the goal posts will keep moving while the technology is on an exponential curve, we are not going to have a magic threshold day when AGI occurred. Instead, in less than 5 years, well before 2030, we are going to be looking back in amazement wondering when we passed the threshold? But make no mistake, that’s going to happen and we don’t have an on/off switch to stop it when it does. This is both exciting and scary.

Dial tone

Leave a reply

When was the last time you heard a dial tone? When was the last time you used a rotary phone? Or a pay phone?

I tend to use Siri a lot. I don’t go to my phone app, I just ask Siri to call the number for me. And I call people a lot less than I text them.

We’ve come a long way from a family phone being connected to our wall, with a long cord attached to the handset. We’ve moved the phone from a side character to the protagonist. From an item to occasionally connect to one person to a tool of constant contact to the world.

Along the way, the idea of the dial tone disappeared. What else has disappeared?

In the middle

Leave a reply

I think that a robust and healthy middle class is essential to maintain a vibrant society. But what I see in the world is an increasing gap between the wealthy and an ever larger group of people living in debt and/or from paycheque to paycheque. The (not so) middle class now might still go on a family vacation, and spend money on restaurants, but they are not saving money for the future… they simply can’t do what the middle class of the past did.

A mortgage isn’t to be paid off, it’s something to continue to manage during retirement. Downsizing isn’t a choice to be made, it’s a survival necessity. Working part time during retirement isn’t a way to keep busy, it’s s necessity to make ends meet.

I grew up in a world where I believed I would do better than my parents did. Kids today doubt they will ever own a place like their parents, and many don’t believe they’ll ever own a house. Renting and perhaps owning a small condo one day are all they aspire to. Not because they don’t want more, but because more seems too costly and out of reach.

Then I see the world of AI and robotics we are heading into and I wonder if initially things won’t get worse before they get better? Why hire a dozen programmers, just hire two exceptional ones and they are the quality control for AI agents write most of your code. Why hire a team of writers when one talented writer can edit the writing done by AI? Why hire factory workers that need lunch breaks and are more susceptible to making errors than a team of robots? While some jobs are likely to stick around for a while like trades, childcare, and people in certain medical fields, other jobs will diminish and even disappear.

I don’t think a robot is going to wanted to provide a pregnancy ultrasound any time soon, but AI will analyze that ultrasound better than any human professional. A robot might assist in laying electrical wire at a construction site, but it will still be a human serving you when you can’t figure out most electrical issues that you have in your home. It will still be a human who you call to figure out how to fix your leaky roof or toilet; a human who repairs your broken dishwasher or dryer. These jobs are safe for a while.

But I won’t want my next doctor to be diagnosing me without the aid and assistance of AI. And I would prefer AI to analyze my medical data. I will also prefer the more affordable products created by AI manufacturing. The list goes on and on as I look to where I will both see and want to see AI and robotics aiding me.

And what does this do to the working middle class? How do we tax AI and robots, to help replace the taxation of lost jobs? What do we do about increased unemployment as jobs for (former middle class) humans slowly disappear?

Will we have universal basic income? Will this be enough? What will the middle class look like in 10 or 20 years?

There is no doubt that we are heading into interesting times. The question is, will these disruptions cause upheaval? Will these disruptions widen the wealth gap? Will robotics and AI create more opportunities or more disparity? What will become of our middle class… a class of people necessary to maintain a robust and healthy society?

Promise and Doom

2 Replies

I see both promise and doom in the near future. Current advances in technology are incredible, and we will see amazing new insights and discoveries in the coming years. I’m excited to see what problems AI will solve. I’m thrilled about what’s happening to preserve not just life, but healthy life, as I approach my older years. I look forward to a world where many issues like hunger and disease have ever-improving positive outcomes. And yet, I’m scared.

I also see the end of civilized society. I see the threat of near extinction. I see a world destroyed by the very technologies that hold so much promise. As a case in point, see the article, “‘Unprecedented risk’ to life on Earth: Scientists call for halt on ‘mirror life’ microbe research”.

We are already playing with technology that has the potential to “put humans, animals and plants at risk of lethal infections.” What scares me most is the word I chose to start that sentence with, ‘We’. The proverbial ‘we’ right now are top scientists. But a decade, maybe two decades from now that ‘we’ could include an angry, disenfranchised, and disillusioned 22 year old… using an uncensored AI to intentionally develop (or rather synthetically design) a bacteria or a virus that could spread faster than any plague that humans have ever faced. Not a top researcher, not a university trained scientist, a regular ‘Joe’ who has decided at a young age that the world isn’t giving him what he deserves and decides to be vengeful on an epic scale.

The same thing that excites me about technological advancement also scares me… and it’s the power of individuals to impact our future. We all know the names of some great thinkers: Galileo, Newton, Curie, Tesla, and Einstein as incredible scientists that transformed the way we think of the world. People like them are rare, and have had lasting influence on the way we think of the world. For every one of them there are millions, maybe billions of bright thinkers for whom we know nothing.

I don’t fear the famous scientist, I fear the rogue, unhappy misfit who uses incredible technological advancements for nefarious reasons. The same technology that can make our lives easier, and create tremendous good in the world, can also be used with bad intentions. But there are differences between someone using a revolver for bad reasons and someone using a nuclear bomb for bad reasons. The problem we face in the future is that access to the equivalent harm of a nuclear bomb (or worse) will be something more and more people have access to. I don’t think this is something we can stop, and so as amazing as the technology is that we see today, my fear is that it could also be what leads to our demise as a species.

The cat’s out of the bag

Leave a reply

I find it mind boggling that just 5 years ago the big AI debate was whether we would let AI out in the wild or not? The idea was, AI would be sort of ‘boxed’ and within our ability to ‘contain’… but we have somehow decided to just bypass this question and set AI free.

Here is a trifecta of things that tell me the cat is out of the bag.

  1. NVIDIA puts out the Jetson Orin Nano. A tiny AI that doesn’t need to be connected to the cloud.
  2. Robots like Optimus from Tesla are already being sold.
  3. AI’s are proving that they can self replicate.

That’s it. That’s all. Just extrapolate what you want to from these three ‘independent’ developments. Put them together, stir in 5 years of technological advancement. Add a good dose of open source access and think about what’s possible… and beyond possible to contain.

Exciting, and quite honestly, scary!

A prediction nears reality

Leave a reply

Andrew Wilkinson said on X:

“Just watched my 5-year-old son chat with ChatGPT advanced voice mode for over 45 minutes.

It started with a question about how cars were made.

It explained it in a way that he could understand.

He started peppering it with questions.

Then he told it about his teacher, and that he was learning to count.

ChatGPT started quizzing him on counting, and egging him on, making it into a game.

He was laughing and having a blast, and it (obviously) never lost patience with him.

I think this is going to be revolutionary. The essentially free, infinitely patient, super genius teacher that calibrates itself perfectly to your kid’s learning style and pace.

Excited about the future.”

– – –

I remember visiting my uncle back when I was in university. The year was somewhere around 1988-90. So, at least 34 years ago. We were talking about the future and Joe Truss explained to me what learning would be like in the coming age of computers.

He said, (loosely paraphrased, this was a long time ago):

‘In the future we will have virtual teachers that will be able to teach us exactly what we want to know in exactly the format we need to learn best. You want to learn about relativity? How about learning from Einstein himself? You’ll see him in front of you like he is real. And he will not just lecture you, he will react to your questions and even bio-feedback. You look puzzled, he might ask a question. He sees you looking up and to the left, which he knows means you are trying to visualize something, and so he changes his lesson to provide an image. He will be a teacher personalized to any and all of your learning needs.’

We aren’t quite there yet, but the exchange Andrew Wilkinson’s son had with ChatGPT, and the work being done in virtual and augmented reality, suggest that Joe’s prediction is finally coming into being.

I too am excited about the future, and more specifically, the future of learning.

How good, how soon?

Leave a reply

I am still a little freaked out by how good the Google NotebookLM’s AI ‘Deep dive conversations’ are. The conversations are so convincing. The little touches it adds, like extended pauses after words like ‘and’ are an excellent example of this.

In the one created for my blog, the male voice asked, “It actually reminds me, you ever read Atomic Habits by James Clear?” And the female voice’s response is, “I haven’t. No.”

Think about what’s happening here in order to continue the conversation in a genuine way. The male voice can now make a point and provide the female voice ‘new’, previously unknown information. But this whole conversation is actually developed by a single AI.

How soon before you have an entire conversation with a customer service representative oblivious to the fact that you are actually talking to an AI? Watch a newscast or a movie unaware that the people you are watching are not really people?

I shared close to 2,000 blog posts I’ve written into the notebook, if I shared my podcasts too and it replicated my voice, I wonder how long it will be before a digital me could be set to write my posts then simultaneously do live readings of them on my blog? Writing and sounding just like me… without me having to do it!

As a scary extension of this, could I learn something from the new content that it produces? Could I gain insights from the digital me that I would struggle to come up with myself?

This is just the beginning. How much of the internet is going to end up being AI generated and filled with AI reactions and responses to other AI’s? And how much longer after that before we notice?

A (creepy) digital friend

Leave a reply

What is Friend? Watch this reveal trailer.

No matter how I look at it, this feels creepy and dystopian. Even when I think of positive things, like perhaps helping someone with special needs, or emotional support for someone with anorexia, the idea of this all-seeing AI friend seems off putting.

Even this advertising doesn’t resonate well with me. In the scene with the guys playing video games, the boy wants to check in with his digital friend rather than pay attention to his friends in the room. And in the final scene with the girl and boy on the roof, I thought at first the girl was candidly trying to take a photo of the boy, but then realized she was just fighting the urge to converse with the AI friend. Either of those scenarios feels like she has replaced a phone distraction with a more present and more engaging distraction… from life.

There are a lot of new artificial intelligence tools that are on their way, and I’m excited about the possibilities, but this one has a high creep factor that doesn’t seem to me like it’s adding the value I think it intends to.