In this three-part series, Damian Davies, Head of Engagement at The Timebank, will explore the rapidly evolving world of AI. He’ll begin with an introduction to AI’s foundations and its applications in business, followed by a look at the risks and ethical considerations it raises. The series will conclude by examining the opportunities AI presents and the tools driving innovation.
In this first article, Damian explores how AI has developed into two core areas—data and language—and what this means for businesses today. Drawing on insights from The Timebank’s Engine Room, he’ll break down key concepts and share practical ways to start engaging with AI.
Tomorrow’s World, Today’s AI
I used to love a TV show called ‘Tomorrow’s World’.
For the young uns (or those who can’t work out what it is from its name), this was a TV show that predicted the world of tomorrow by looking at gadgets or innovations that were being developed.
It ran from 1965 to 2003 and, like Doctor Who or James Bond, your favourite era was probably influenced by your age when you discovered it.
For me, it was the Judith Hann, James Burke and Keith Chegwin’s missus era.
To be fair, it was the era with the best theme tune, a cold, synth-pop track, which is probably the reason I went on to spend so many weekend nights in dark rooms listening to repetitive beats.
The programme’s demise happened before the rise in handheld technology, so it missed the massive revolution we are going through now.
The revolution of Artificial Intelligence.
In a series of three articles, I am going to try to unpack some of the basics behind AI and also looking how it can and will be deployed in every financial planning and advice business.
I am not an expert, but I am fascinated by AI, so I am just going to share what I have learned and hopefully give you enough information to start exploring it for yourself. I may even use AI in producing some of this text!
Background
AI was entirely philosophical until Alan Turing laid the groundwork for AI as a computing concept with the idea that a machine could simulate any other machine’s logic. This was further advanced with his famous 1950 paper, “Computing Machinery and Intelligence,” in which he posed the question, “Can machines think?”.
This introduced the Turing Test, a method to evaluate a machine’s ability to exhibit intelligent behaviour indistinguishable from that of a human.
The term Artificial Intelligence itself was coined in 1955 by American computer scientist John McCarthy, and The 1956 Dartmouth Conference is often considered the birth of AI as an academic field. Researchers like McCarthy, Marvin Minsky, Nathaniel Rochester, and Claude Shannon gathered there to discuss whether machines could be made to simulate every aspect of human intelligence.
In the decades that followed, AI research moved through periods of optimism and frustration. Early AI programs, such as the Logic Theorist (1956) and General Problem Solver (1959), demonstrated that machines could solve problems by following logical rules. However, progress was slow, and AI development faced a significant setback in the 1970s due to limited computing power and overly ambitious expectations.
The resurgence of AI came in the 1980s with the advent of expert systems, which were designed to mimic the decision-making abilities of human experts. The real leap forward in AI, however, came in the 2000s with the rise of machine learning.
This is a subfield of AI that focuses on developing algorithms that allow machines to learn from data.
This shift paved the way for deep learning and modern AI breakthroughs, which has brought us to the OpenAI era.
As you can see, AI has followed a pretty typical ‘branch’ form of evolution. For the nerds, it’s called ‘phylogenics’, which means lineal descendants follow a common ancestor.
This evolution means AI is being developed and deployed in two areas.
DATA – analysing and reviewing lots of data and presenting the analysis in a way to suit the user.
LANGUAGE – analysing and reviewing lots of words and presenting the analysis in a way to suit the user.
The only problem is that AI is like that really brainy but totally thick person. I know that seems counterintuitive, but you will have met people like this. They are frighteningly clever, almost too clever, as they trip over their own toes or have no room for common sense.
That’s AI. You can’t expect AI to just plug into your life and do everything for you. It needs to be trained.
Once it is trained, however, it is genuinely astonishing what it can achieve.
That’s what is happening now when someone tries to flog you an AI system.
It means that we have harnessed AI as either a data or language tool and taught it to do something.
The quality of that tool is down to the quality of the training the AI has had.
The truth is some tools are good, some are bad, and some solve a problem that doesn’t exist!
At The Timebank, we have been exploring AI tools for a couple of years now.
In fact, we have set up a lab called The Engine Room, where we are testing various tools on live cases. If you are intrigued, you are welcome to log an interest here.
In the next article, I will explore some of the risks of using AI and how to mitigate them. After that, the final article in the series will unpack some of the tools you might want to start using.
