AI: The Nuances

Would AI kill us? Take our jobs? Or Make our lives easier?

Happy Monday!

Artificial Intelligence, AI, stands at the forefront of today's rapidly advancing technologies, with the promise to reshape human life in ways previously thought impossible. However, like any powerful tool, AI brings with it a host of potential benefits, risks, and complexities that need careful consideration.

Here is the third in a series of four on my fully-fledged thoughts about Artificial intelligence.

If you missed the last edition (AI: The Boom ), here's a summary:

• Machine Learning is a part of AI that lets systems learn from data instead of being explicitly programmed. It figures out patterns and predictions by analyzing data, as seen in the example of a football AI. Unlike traditional programming, it derives its own rules based on inputs, leading to creative and unexpected outcomes.

• The rise of Machine Learning is due to improved computing power and the shift from CPUs to GPUs for AI tasks. GPUs can handle multiple tasks at once, boosting efficiency and enabling exponential growth in AI capabilities, even defying Moore's Law. This power has incredible potential but also raises several concerns.

TL;DR - AI: The Nuances

• AI, like "The Monkey's Paw" story, can fulfil requests with unintended consequences. It has immense potential in medicine, education, agriculture and more, but its power brings concerns about extinction-level risks.
• An AI may lack an understanding of human nuances, leading to harmful outcomes. The fear is it might excel at precisely what we ask but cause harm.
• While AI can disrupt jobs, history shows it can also create new opportunities, complementing human abilities.

AI: The Nuances

To truly describe the landscape of AI, I like to link it to the tale of "The Monkey's Paw", a short story by English author W. W. Jacobs. The story follows the

To truly describe the landscape of AI, I like to link it to the tale of "The Monkey's Paw", a short story by English author W. W. Jacobs. The story follows the experience of a family - the White family - with a mystical talisman rumoured to grant three wishes.

After obtaining the talisman from a friend, the Whites wish for two hundred pounds, but the windfall comes at a terrible cost — their only son is killed in a workplace accident. Incidentally, they get 200 pounds from their son's workplace as a sort of consolation.

Grieving and desperate, they use their second wish to bring their son back to life, only to be horrified by the twisted and mangled figure that appears at their doorstep. Consumed by fear, they use their final wish to end the horror.

Like the monkey's paw, AI can fulfil EXACTLY what we ask of it, even if it's nowhere near the outcome we want.

The Good: Beyond Mundane Tasks

The potential of AI is enormous, but its true value doesn't solely lie in its ability to complete mundane tasks. Rather, it lies in its potential to catapult us beyond our current limitations.

Consider the trajectory of human progress, starting from the discovery of agriculture. The emergence of agriculture freed up much of our time as hunter-gatherers, allowing us to delve into more intricate fields like astrology and more.

Notwithstanding, AI wields the promise of being the largest breakthrough man has had in the last century, surpassing the following:

- Electricity: which made it possible to run machines more quickly and efficiently

- Rail: which connected us in ways previously unimagined

- Information Technology: which brought an unprecedented level of information to our fingertips.

AI is a vehicle that can leapfrog humanity towards achievements we couldn't fathom accomplishing in (arguably) ALL fields. Here's proof:

1. Medicine and Healthcare: AI has been used in medical diagnosis and treatment by analysing vast amounts of medical data such as images, genomic sequences, and patient records, to identify patterns and predict disease outcomes. For instance, AI-powered diagnostic tools can detect abnormalities in medical images like X-rays and MRIs more accurately than human experts.
2. Education: In education, AI can learn to personalise learning experiences. Schools and educators are using adaptive learning platforms to tailor educational content based on individual needs and pace, improving engagement and understanding. Khan Academy's Khanmigo is a classic example of how AI can be used to provide world-class AI education to anyone across the globe!
3. Agriculture: In agriculture, AI is helping farmers to improve their efficiency and the sustainability of their practices. Farmers are using AI and satellite data to optimise planning, irrigation and fertilisation - improving crop yields and reducing wastage. AI-driven pest detection systems are also helping farmers to identify and address infestations early, minimising crop loss.

The possibilities with AI are just... endless and that is the beauty of it. It often unveils uncharted territories and novel possibilities.

Yet, we don't know the full scope of what it'll help us do or unlock. And that is the scary bit. We can see the results it produces, but comprehending its processes remains elusive.

The Bad: Unintended Consequences

Bill Gates, Sam Altman and 100s of tech CEOs recently signed a profound 22-word statement that read:

"Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war."

This call to action rightly underscores the growing concern surrounding AI's capabilities. Come to think of it, the threat from AI is being compared to destruction by Pandemics and Nuclear War. Scary! The fear, like the tale of the Monkey's Paw, is that AI would do exactly what we tell it to do, and not what we want it to do. AI's rapid evolution sparks fears that it might acquire a level of power that could jeopardize humanity.

But while an AI may appear to be human and act like one, it is not. It doesn't want anything, it can not feel and it has neither intentions nor emotions. An AI is just a collection of algorithms choosing optimal outcomes from a sea of possibilities it has learned. So our fear is in the fact that it might choose a possibility that would cause us significant harm, without an understanding of the nuances that make up our world.

Let's use football as an example. You may train an AI to win you a match, and it would score tons of goals, yes, but leave all your players injured for the season. It may choose to optimise for winning a match (that may not even be that important) by playing in a way that gets 8/11 players injured.

We fear, ironically, that Artificial Intelligence would become so good at doing exactly what we ask.

Would AI Steal Our Jobs?

Throughout humanity's existence, the emergence of new technologies has always threatened existing jobs, and that is true for Artificial Intelligence. The rise of AI has caused a lot of scepticism about how many jobs will be lost, and the unemployment it will bring. To this, I like to answer with the rise of the ATM in the banking sector.

Did you know that ironically, the advent of the ATM created more jobs for tellers in the banking industry? While human tellers were no longer limited to spending a large chunk of their time cashing in and out money for clients, they had time to provide better customer service and more. That way, banks' operating costs fell and they had more resources to open more branches and reach more people, thereby hiring more tellers.

Al can provide enormous value by complementing both labour and physical capital in ways that enhance productivity. By automating repetitive and mundane tasks, employees can now focus on high-value-added tasks and more advanced cognitive functions of their jobs.

For example, instead of having accountants spend so much time compiling data and crunching numbers, we can have them focus on communication, relationships, and better assessing risk using higher quality data. Al will have a significant impact by enabling faster and more profound progress in almost any field where intelligence (human or artificial) has a role to play.

Such an exciting future!

But what does it mean for Africa, the continent that's been constantly stuck playing catchup over the past few centuries? This and many more are answered in AI: The African Side next week.

Ponder Goodies

Top AI tools to take advantage of:

Category: Marketing

• Jasper.ai : Jasper.ai ’s easy-to-use interface can help you generate ideas, craft first drafts, and edit final posts.
• Hubspot : Hubspot’s content assistant is a suite of free, AI-powered features that help you plan, create, and share remarkable content quickly.
• Adverity : Centralize all marketing data from various sources, including campaigns across all channels, where it can be easily analyzed.
• Zapier : Scale your business by automating time-intensive tasks that could include sending contacts to email lists and kicking off important workflows.
• Brand24 : Sift through massive amounts of information generated on public platforms, whether that’s social media or forums and analyze everything from mentions to the volume of discussion and overall brand sentiment.
• Semrush : Generate content and ensure your writing is optimized for SEO.

About Ponder

Ponder, my blog, is my attempt to share my random thoughts with you to inspire you, cause you to challenge your thinking and even make you laugh (Hopefully).

With ponder, you get to explore Minimalism, Christianity, Science, Graphic Design, an annoying appreciation of detail and out-of-the-box thinking. You get to explore all of these from the perspective of an African.

There’s more on my website: joshwordey.org/blog