Category: Uncategorized

AI – Our New Overlord?

Posted on by Steve Sammartino

Click below to listen to Steve read this post (11 min audio)

I remember pundits exclaiming the personal computer, the internet and the smart phone would change the world. And they did….

But, this is the first time I can remember new tech being discussed as an existential threat to our species.

Recent developments in Large Language Models such as ChatGPT are significant enough that some of the world’s most qualified researchers are making a plea for us to stop before things spiral out of hand.

A Rare Event

An Open Letter published by The Future of Life Institute requested we put the brakes on all AI research. This is a rare event. From recent memory, we’ve only ever had moratorium on human cloning, and germ-line modifation.

This time, they are less worried about the modification of our species and more about creating another ‘species’. One which superior to us in every measurable way.

As a long-time technologist, I think it would be prudent to take heed of the letter. But first, we all need to understand the different types of AI, what they mean and what they can do, as most people lack this context.

Today’s post is an important explainer.

Types of AI

At its simplest, Artificial Intelligence is a field which combines computer science and robust datasets to enable human like problem-solving. AI is achieved by creating algorithms that are capable of learning from and adapting to data and experiences.

It’s different to industrial machinery, as the software uses AI to ‘think’, like a human and perform tasks on its own. The AI chooses what to do, based on a set of instructions. A machine simply performs a pre-determined task upon instruction.

There are 3 clear types of AI’s:

  • Artificial Narrow Intelligence (ANI)
  • Artificial General Intelligence (AGI)
  • Artificial Super Intelligence (ASI)

Artificial Narrow Intelligence: ANI represents the large majority of AI’s we use in modern society. These are generally rule-based AI’s which decide something based on an ‘If This, Then That’ Protocol (IFTTT). A simple example of this is would be Google maps which generate for you the quickest way to get somewhere based on distance, traffic conditions and estimated traffic conditions during the journey. A self-drive car also operates under this doctrine. We could even add image recognition, and deep fakes to this list. They are described as narrow because they typically operate in a vertical. The only context they have is within a specific framework. These AI’s do improve, iterate and learn, as their dataset gets bigger. But they are limited to a specific realm. While most of these ANI’s outperform humans, they lack nuance, are imperfect and can’t really be applied to different tasks.

Artificial General Intelligence: ChatGPT falls into this description and is the reason why people are becoming concerned. AGI’s take intelligence and make it horizontal, meaning they can be applied to a number of, and potentially unlimited, contexts – much like biological beings. An AGI is capable of performing any intellectual task a human can. Such systems can reason, learn, plan, understand natural language, recognise patterns, and solve problems. This is what Large language Models can do, and the reason they can do it is that they’ve learned via the entire gamut of human experience, in what is our species’ killer app – Language. All they need is the correct input. AGI is intended to be a more flexible and adaptable form of AI. It can understand new situations and environments. AGI systems are capable of generalising knowledge and taking skills from one domain and applying it to another. Already, ChatGPT has shown a number of emergent capabilities and intelligences it was not designed for. These were surprises to the developers themselves. The fear is this will lead to self-awareness.

Artificial Super Intelligence: ASI’s are different, and the major risk we face as this ‘unauthorised’ AI experiment continues. It refers to a hypothetical form of artificial intelligence that surpasses human intelligence in every way and is capable of designing and improving its own systems, even beyond human comprehension. It has the potential to develop its own objectives, agendas, and organise the factors of production for its own purposes. It can go beyond instruction and become a new form of living entity, which can spawn new improved versions of itself, and even organise hardware to build out any physical manifestations of itself, which it may require to perform tasks against its own desires – whatever they may be.

ASI is considered the ultimate form of artificial intelligence, and could potentially revolutionize, or end civilisation as we know it.

Didn’t mean to scare you.

Your business should be scared of AI – only if it doesn’t get me in to deliver my mind blowing new Keynote on AI  – it’s a game changer.

The Singularity

The idea of the Singularity, as proposed by Ray Kurzweil, is based on the concept that once ASI is achieved, it would rapidly accelerate technological progress and fundamentally transform human civilization. According to Kurzweil, the Singularity would mark the point at which technology becomes so advanced that it is no longer possible to predict or comprehend the future. Kurzweil believes that this is both inevitable and irreversible — He currently predicts this will happen by 2045.

Externalities

Humans are typically slow to respond to externalities. This was true with climate and fossil fuels. But we didn’t realise carbon emission posed a threat until fossil fuels were ensconced in the modern economy. This is not one of those times. If Kurzweil is right, global temperature could be the least of our worries.

The biggest challenge is that we live in an era of democratised technology. No one needs government approval to commence their own AI development lab. No oner needs Manhattan Project style budgets. But the implications could be ‘nuclear.’ In addition, our geo-political environment is one in which we can be sure the Chinese Communist Party won’t press pause on AI, even if democratic countries do. 

While it could be that we never generate an ASI (I’m still waiting for my fully autonomous vehicle) this is a time when prudence is desirable. As the Open Letter mentions, the last thing we want is unelected tech leaders putting our civilisation on the line. What I’d much rather is courageous leaders using the powers they’ve been granted to make decisions no individual can.

—

Keep thinking,

Steve.

The Age of Viral Finance

Posted on by Steve Sammartino

Click below to listen to Steve read this post (5 min audio)

In a digital world, if we collectively believe in an outcome, it’s very likely to occur.

This is not a new concept. Behavioural Economics says that the economy will respond to the future conditions people believe will transpire. Because they expect it, they make it so. Downturns can become inevitable, just by thinking one might happen.

But it isn’t just memes and videos that can go viral. Our financial markets, which are fully digital, have also become susceptible to virality. An idea can become true, a market can crash or a bank can fail, simply because we think it might happen. The idea becomes the truth. This just happened with Silicon Valley Bank. Their story is an allegory for our modern world.

World’s First Viral Bank Failure

Silicon Valley Bank (SVB) is a very important bank to the startup and tech eco system in the USA. In 2021 when times were good and money was flowing, they were filling up with deposits. Just like any bank, they wanted to put those deposits to best use. So they bought a large amount of long term bonds at the then interest rates at a little over 1% per annum. This was a quasi bet that interest rates wouldn’t change much. Granted, they had been very low for around a decade. Then in 2023, the interest rate goes from near zero, to 5%. This means that SVB has all these unrealised losses on their books. If they had to sell them in the short run, they’d be in some trouble. The reason is that the new higher interest rates, make the bonds worth less, around 95 cents on the dollar. At the same time, the startup eco system was simply spending their capital and not raising more, because financial markets were tightening. And this started to create a bit of a crunch.

Then, in February, a tech blogger named Byrne Hobart wrote a post proclaiming the SVB was functionally insolvent. It actually wasn’t – all banks have less money than they take in. (In Australia, banks only require 17.5% of deposits on hand). The rest they loan out. As you can imagine, this blog post, raised many eyebrows in Silicon Valley. People started to worry. Then, within a couple of weeks, various venture capitalist group chats, all started to send messages around, advising their portfolio companies to withdraw their money from SVB. And because everything is digital, this happened very quickly. As would be the case with any bank, if everyone wanted their money back immediately, they wouldn’t be able to do it. Within weeks SVB had to shut its doors. It was the fastest Bank Run in US history.

This was the worlds first ever Viral Bank Failure

In the same way that a tweet can go viral, so can the idea that a bank might fail, go viral. SVB became insolvent, because people thought it would become insolvent. A bank which was valued at over US$40 billion a few months back, no longer exists.

Learn how fast AI is changing our world – get me in to deliver my mind blowing new Keynote on AI  at your next event – Time waits for no one.…

Collective Digital Reality

It’s as if the digitisation creates a natural gravity. It empowers ideas which are popular to win, regardless of their veracity.

As a species, collective thought is more important than reality. Once we believe something strongly enough, it becomes a reality. We do this with our gods, our currencies and our economies. We make myths a reality.

The only challenge, is that in a hyper connected world, dangerous ideas can become real – real quick, regardless of what the science says.

—

Keep Thinking,

Steve.

Your Own Personal Jarvis

Posted on by Steve Sammartino

Click below to listen to Steve read this post (6 min audio)

The AI’s such as ChatGPT, are taking the world by storm. While these are ‘institutional technologies’ we’ve been granted access to, they always trickle down to the domestic level. What’s next will change our personal lives beyond belief.

Within 3-5 years we’ll all have personal AI’s. Your own Jarvis – guiding your life through every – single – moment. A brain extending AI which always serves your needs – not that of a corporation.

It will be more powerful than ChatGPT – because it will be trained on your personal data set.

Teaching Your Own AI 

At the moment, we are training the AI’s as we interact with them. Firstly, through the entire gamut of human experience which has been published on the web for the past 3 decades. And secondly, via our interactions with the AI’s as we query them. These provide valuable feedback loops. Every query and prompt we put into an AI like ChatGPT, makes the AI smarter, as does our revised prompts after each moment of data generation.

It’s a bit like teaching children. The AI observes. Responds, and gets told again – “No, not like that, like this!”. This all has a sense of biomimicry about it.

Once these web enabled AI’s are sufficiently trained on large datasets, they’ll be able to function on much smaller, even micro datasets such as those on our personal devices.

Our own devices are filled with everything that matters to us. We get a new device and transfer that data to the next, and the next, and so on. Most people have a complicated combination of data in cloud storage and on hard drives across their PCs and smart phones. This is increasingly dense and difficult to navigate and retrieve data from.

This data isn’t trivial – it is a mirror of our modern lives.

Everything we think, say and do. Everywhere we go and who we are with is recorded on our devices. Think of all our pictures, our videos, our finances, our voices. This will be the next bastion of AI. Just like the computer did, the AI will become personal. We’ll plug in an AI to our devices, it will crawl through all of the data, and take what it has learned from the internet, and be our own generative AI – We’ll all have our own Jarvis. It will think, act and create like us – but be a smarter version.

Democratised Intelligence

IQ is over. There will be no intelligence advantage for anyone. Just like no one really gets a better smart phone. It’s equalised. What will become important however, is the data our own AI will learns from. Those with less personal data and records of their work and their life will be at a disadvantage, because the AI won’t have anything to learn from. This is already evident with ChatGPT.

AI & Personal Brands

If personal brand wasn’t already important, AI just gave everyone who has a digital footprint an advantage – you can now ask it to do work in the style of “You” – but only if the investment has already been made and well published. Personal branding has just moved beyond selling yourself and even ego. To now informing a vital modern day tool – your own AI.

I have written more than 3,000,000 words on this blog over the past 20 years…. and published large amounts of public data, under my name Steve Sammartino. So when I ask ChatGPT to create something in the style of ‘Steve Sammartino’ it really sounds quite like I wrote it. You can see this prompt I did as a test on ChatGPT-4:

Chat GPT4 Does Steve Sammartino!

You can read what it came up with here. I’d love to know if you think it sounds like me. Just reply email to this post (it comes directly to me).

You’ll also love my new keynote – get me in to deliver my mind blowing new Keynote on AI  at your next event – Don’t regret getting left behind in the biggest change in more than a decade.

While Siri isn’t great, I would be flummoxed if Apple doesn’t arrive with the worlds first personal AI in the the next few years. And if they do, it will change the world more than the iPhone did.

—

Keep Thinking,

Steve Sammartino

AI Generated Entertainment

Posted on by Steve Sammartino

The image above is from a show called ‘Nothing Forever’. The show is AI-generated version of Seinfeld.

It has been streaming on Twitch 24/7 since mid-December. It tells the “story” – if you can call it that – of four characters, Larry, Fred, Yvonne and Kakler, who look like what would happen if Jerry, George, Elaine and Kramer were sucked into an 8-bit video game from 1990. You can watch season 1 here.

While Seinfeld ended in 1998, it’s been the subject of a zillion mashups since then, including the great Twitter feed called Modern Seinfeld, a TikTok actor doing his own spin on it and even some Hip Hop Tv fusions. But at the top of the list we have to put Nothing Forever, because we’ll look back on this as the first of something which changes entertainment forever.

AI’s will win an EMMY

This is actually some cutting edge stuff, because all of it was creating by AI and Machine Learning. The script, the voices, the computer imagery, the movement, the laugh track and even the music is very Seinfeld-esque. It runs forever…. it doesn’t end, it just keeps on going. It is running right now here, and it has 200K followers.

At the moment the show is super glitchy. The dialogue is mostly weird, the characters bump into each other, and the images are super lo-fi. But it occasionally has some genuinely funny moments. It is the just the seed of a much bigger movement. Let me explain.

Just like video games, this will improve and eventually evolve into HD video of people which are indistinguishable to the real Seinfeld actors. You won’t be able to tell the difference. The jokes will get funnier. In short, AI will be able to generate an infinite future of Seinfeld, by learning from actual Seinfeld episodes, and combining information from the modern world to make it relevant to today. As the AI improve exponentially, so will the output of shows like this.

Re-Runs Version 2.0

Within a decade you’ll be able to watch AI generated versions of whichever TV show you used to love. Friends, Happy Days, Mash, Game of Thrones, Mad Men, Big Bang Theory…. you name it. You may remember 4 years ago write about you becoming a movie star. This is part of the same movement.

Find out now what world first you can create with AI – get me in to deliver my mind blowing new Keynote on AI  at your next event. You’ll love it.

This is the Future of Entertainment – At first it will be re-runs which are just as good as the original, and eventually entirely new shows will be spawned via AI with no real actors or script writers. They’ll look sound and feel as thought they were created in the traditional way, and we won’t care. Studios will embrace it because it will be cheaper, and viewers as always, just want to be entertained. We’ll develop emotional connections with actors, in Tv shows that don’t actually exist.

AI is changing everything.

—

Keep Thinking,

Steve.

A.I.’s Window of Opportunity

Posted on by Steve Sammartino

Click below to listen to Steve read this post (3 min audio)

Remember this iPhone advertisement? – There’s an App for that? – Of course you do. I’m sure you had a number of App ideas that you never got around to making. (we all did)

Then, you saw Apps that were similar. They validated your idea, and made you kick yourself that you didn’t act quicker. Side note: There are very few things in my career that I wish I waited longer to do!

Don’t wait to get the low down on AI – get me in to deliver my mind blowing new Keynote on AI to your team – it might just provide the seed for something amazing!

A moment in time

Untold billions have been made on the back of the smartphone. Some built reputation, some made money, some got sold to Apple and some, even turned into giant global businesses in their own right. Despite starting inside another corporations eco system, as a simple App.

What is happening right now with Artificial Intelligence is a once in a decade tech shift. One in which new fortunes will be made. We saw this with the World Wide Web, when Social Media emerged, and with the Smartphone. It is now that time for democratised AI. Society is about to change. This chart really displays the moment in time which has just arrived.

Each pink block represents the size of opportunity of each technology epoch. Not only are they bigger, they change the world faster, as represented by the increasing vertical leaps in each block. This is exponential technology in action.

At my keynote speeches recently, I’ve been asked what the biggest opportunities emanating out of AI will be….

This is my answer:

It won’t be what you think. The real opportunity with AI, ChatGPT and friends, isn’t doing what you’ve been doing ‘more efficiently’. It’s not about AI absorbing tasks. It will be an unexpected use of the tools. They’ll be an unlock which becomes obvious in hindsight. If we look back at the Smartphone, the biggest unlock was actually the GPS. The ability to locate things and people invented massive businesses. It made Uber, Airbnb and Dating apps possible. Just think about how weird it would’ve sounded if someone said 20 years ago;

‘I think satellites will have a big impact on relationships and marriages. They’ll disrupt bars and alcohol as a way for people to meet.’

It would sound insane, yet here we are. Likewise, it’s early for AI and creative people will uncover new ways to use the technology which we and the creators of the technology don’t foresee. That is the opportunity. And the only way to find the opportunity is to experiment with the tools and let your imagination run wild. Because leveraging technology is never about understanding how it works, and always about to get it to work for you.

It’s an imagination game…. we’ve been handed a platform, it’s time to experiment.

—

Keep Thinking,

Steve.

AI is Redefining Real

Posted on by Steve Sammartino

No AI’s have been used in the creation of this post. Of course, you can’t really be sure of that. You just have to take my word for it.

If it seems like I’m obsessed with AI at the moment, trust your instincts. It’s a worthy topic because the universe just changed. Last week I gave you a small taste of the tools available. This week we need to consider the implications of the tools. We can start with this image of random people below. And they all have one thing in common. See if you can guess what that is?

None of these people actually exist. They are computer generated photos of imagined people. Well, at least that’s what www.thispersondoesnotexist.com will have me believe. For all I know they took a phone to a BBQ and took some happy snaps.

And that’s one of the key issues we’ll face in the next few years. Knowing if something is real or AI generated. Just think of all the synthetic versions of reality we can already create. Voices, videos, pictures, in fact anything we can create a digital version of, we can now create a fake version of. And it has no noticeable difference. Within a decade we’ll be able to buy soft robots of ourselves. Which look and sound exactly like us. In fact, this has already happened. Hiroshi Ishiguro is a roboticist who makes life like robots, including one of himself.

We may need an entirely new definition of what is real.

Understand how “real” AI is about to get with my mind blowing new Keynote Speech – get your team up to speed – The time is now.

So what should real mean? The original. There is only one. The first. Organic. Not made by machine of software. There is no substitute. Found in nature. Not an imitation. Not made in a factory. All natural ingredients… And who gets to decide?

Or maybe, it’s natural that AI has arrived, because it was created by biological beings. It might just be just a fork in the inevitable evolutionary path we must follow. I’m not exactly sure.

But here is what I know for sure. The only way we’ll be able to tell if something is real of ‘fake’ – is if we are forced to outline it legally. It’s time for law makers to get busy, fast. And while I’m very positive about what emergent AI’s can do to help humanity, I also think humanity deserves to know when they are interacting with each other, or something other.

—

Keep Thinking,

Steve.

Disrupting Google

Posted on by Steve Sammartino

Business disruption is not caused by technology alone. For it to occur we need 2 things to arrive simultaneously.

(1) A new technology

+

(2) A new business model

If we only have one, the incumbents can usually adapt. They can plug the new tech into the existing business model. Or, they can revert the old technology into a new business model.

For example:

The Music Industry had 3 new technologies before they got disrupted. They had the phonograph, the tape and the CD. Each time they sold the new tech in the old business model. It wasn’t until the mp3 arrived until the industry changed. When that happened, the business model shifted with the tech, which resulted in disruption: Napster (stealing music) and Apple iTunes (buying music one song at a time). Then when streaming arrived, a further disruption occurred as both the tech and business model shifted once more. No one buys music, they subscribe to it.

Likewise, when the Airline Industry had low cost airlines arrive. A new business model emerged, but because it was utilising existing technology: planes, airports and booking engines, legacy players could plug in low cost sub-brands. No real industry disruption transpired.

Most Successful Consumer Product Launch in History

Chat GPT is the fastest-growing consumer product in history. It had over a million users in its first week and more than 100 million in two months. Previous technology juggernauts haven’t come close: TikTok took nine months to get to 100 million users, Instagram took nearly three years and Google took nearly two years to reach this milestone. It isn’t just the rapid growth of users of the platform that’s interesting. It’s that it demands a review of internet Search as we know it, how we perform searches literally and the resulting business model which underlies it. It may even redirect us away from advertising and the prevailing surveillance capitalism model.

The technology and business model just changed for search. Sounds crazy to say it, but Google could be in trouble. If there was ever a company which looked dominant and unstoppable mere months ago, it was Alphabet. Their Google search engine commands a 90%-plus share in most of the markets it operates in. Then along came ChatGPT.

Will your company be the Disrupter or the Disrupted with AI ? Get me in to share my mind blowing new Keynote Speech – and win in the new AI era.

Bing v Google

At the moment it looks like Open AI, the developers behind ChatGPT, have everything to gain, but behind the scenes is tech overlord Microsoft. If all goes to plan they could be the unexpected winner in AI, and there are literally trillions of dollars in market capitalisation at stake. Microsoft’s 23 January $10 billion investment in Open AI may well be the tech deal of the century. As a part of it Microsoft will have exclusive access to Open AI’s product suite, and will gain a 49% share of Open AI. However, Open AI will need to give back Microsoft 75% of the profits until Microsoft recoups its initial investment. Microsoft have already plugged ChatGPT into their Bing Search engine, and it is pretty damn good. I’ve switched already. But is isn’t just the product which puts google at risk, it’s the costs and business model.

The cost per ‘prompt’ on ChatGPT is currently around $0.02c. This is vastly more than the $0.00001 per Google search, and probably couldn’t support a pay per click or display advertising model. The recent option to subscribing to ChatGPT for $20 per month is a clue as to where the business model of Generative AI is likely to go – subscription rather than advertising. This would both remove the ‘free rider’ problem, and temptation to compromise product quality to appease the advertising model supporting it. Subscription is also needed because AI is far too expensive per prompt to run a pay per click model. This is a major problem for Google – which people use for free.

The market is likely to bifurcate into two segments: Search (Traditional web links) and Creation (Generative AI).

Think about it – if we shift our search habits to ask questions and getting an actual answer, rather than a page of links and options – the pay per click model could die alongside it. Bing might just become the world’s first Premium Search engine – a pay to play for a different kind of search.

The Code Red which was called in through halls of the GooglePlex hasn’t resulted in anything that seems like a worthy response to ChatGPT. After a failed demo last week of the Google AI chatbot Bard, it lost more than $100 billion in market cap. But I also wonder if the market senses that Google has far more to lose even if (and most likely when) it develops a competitive AI product. 58% percent of Alphabet’s revenue comes from search, which is driven by pay per click advertising, which simply can’t survive with generative AI – there are literally no clicks when you get a direct answer. Currently Microsoft only generates 5% of its revenue from Bing pay per click advertising. In real terms, it has a potential ten-fold search revenue upside, with near zero downside all the while potentially adding a new weapon to its already strong enterprise offers of Windows, Office and Azure. AI inside your own laptop, generating answers from your own personal data. That would be super powerful, personally and at an enterprise level.

Just when we thought we thought a one tech firm could never be usurped, a new technology comes along which potentially changes everything.

– – –

Keep Thinking,

Steve