Introduction to Artificial Intelligence (AI)

Artificial Intelligence, often shortened to AI, is a branch of computer science that focuses on creating and utilising computer systems capable of performing tasks that typically require human intelligence. These tasks include

• • learning from experience
  • understanding natural language
  • recognising patterns
  • making decisions

AI is a type of technology that allows computers to learn and make decisions like humans. 

AI systems aim to simulate human intelligence by

• • Acquiring knowledge (gathering information and understanding its application).
  • Reasoning (using the information to make conclusions, either approximate or definite).
  • Correcting themselves when they make mistakes.

Misconceptions About AI

Before going further, it is worth clarifying what AI is not. Here are some of the most common misconceptions about artificial intelligence.

• • AI is a single technology: On the contrary, AI has a vast array of technologies under its umbrella. There are many different types of AI system, some for general use, and others for specific use e.g. medical diagnosis, financial trading, robotics etc. In addition, many different companies have created their own AI platforms e.g. OpenAI, Google, Meta etc. There are also thousands of tools and apps that use different AI systems.
  • AI is all-knowing: Another misconception is that AI systems have unlimited knowledge and can provide accurate answers to any question. In reality, AI systems are limited by the data they have been trained on, and may not always have the correct or complete information.

To understand AI it is also important to understand what it is not

• •  AI is infallible: Some people believe that AI systems are always right and do not make mistakes. However, just like any other technology, AI systems can have errors and biases that need to be addressed and corrected.
  • AI is only for large companies: Many small businesses may think that AI is only accessible to big companies with large budgets. In reality, small businesses can benefit hugely from using AI, and can gain the ability to compete with larger companies more cost effectively.

AI Technologies

Here’s a brief overview of some AI technologies:

• • Machine Learning: The ability of machines to automatically learn from experience, without being explicitly programmed.
  • Natural Language Processing: Understanding and interpreting human language.
  • Robotics: Designing, constructing, operating, and applying robots.
  • Speech Recognition: Translating spoken language into written text.
  • Computer Vision: The ability of machines to see and understand the world around them e.g. facial recognition and medical image analysis.

Understanding Machine Learning

Artificial Intelligence, as we’ve learned, is a broad field which uses multiple different technologies. One of these technologies that significantly contributes to AI is Machine Learning. To put it simply, machine learning is the process by which computers improve their performance without explicit programming. They achieve this through algorithms (sets of rules or instructions) that can learn and adapt from data.

In simple terms, machine learning is a way for computers to learn from data, much like how humans learn from experience.

Machine learning algorithms play a central role in AI. They are the driving force behind the ability of an AI system to learn, adapt, and improve over time. Instead of being explicitly programmed to carry out specific tasks, machine learning algorithms allow systems to analyse data, detect patterns, make decisions, and even predict future outcomes.

Machine Learning algorithms are often categorised into three types –

• • Supervised Learning
  • Unsupervised Learning
  • Reinforcement Learning

Supervised Learning

Supervised learning is a type of machine learning where the algorithm learns from labelled data. It requires input data and corresponding output data to train the model. Here are some easy-to-understand examples of supervised learning:

• • Email Spam Classification: In this example, the algorithm is trained to classify emails as either spam or not spam. The input data consists of various features of an email (such as subject line, sender, message text, attachments etc.), while the corresponding output data provides labels indicating whether each email is spam or not. By analysing these labelled examples, the algorithm can learn patterns and make predictions on new, unseen emails.
  • Fraud detection: Machine learning can be used to detect fraudulent transactions in real time. This is done by training an algorithm on a dataset of known fraudulent transactions. The algorithm then learns to identify patterns that are associated with fraud. This information can then be used to flag potential fraudulent transactions before they are completed.
  • Credit scoring: Supervised learning can be used to train models to predict the likelihood of a person defaulting on a loan. This is done by providing the model with a dataset of people who have defaulted on loans and people who have not, and the model learns to identify the features that are associated with defaulting.

These examples demonstrate how supervised learning algorithms can be used to solve real-world problems by learning from labelled data to make predictions or classifications.

Unsupervised Learning

Unsupervised learning is a type of machine learning where the algorithm learns from unlabelled data. This means that the data does not have any pre-defined labels or categories. Thus, the machine has no way of knowing what the data represents. In effect, it is just a collection of raw bits and bytes. The algorithm learns to identify patterns and structures in the data, without having any prior knowledge of the data.

Here are some easy-to-understand examples of unsupervised learning:

• • Music genre classification: Unsupervised learning can be used to classify music into different genres based on their audio features. This can be useful for tasks such as music recommendation and music discovery. For example, a streaming service might use unsupervised learning to classify music into different genres based on the songs’ tempo, instrumentation and other audio features. This would allow the service to recommend music to users based on their listening preferences.
  • Image Recognition: Unsupervised learning can be used to recognise objects in images. This can be helpful for businesses to organise their images or to identify different types of objects. For example, a security company might use unsupervised learning to recognise objects in security footage, so that they can identify potential threats.
  • Cybersecurity: Unsupervised learning can be used to detect malicious activity. This can be useful for tasks such as intrusion detection and malware detection. For example, a company might use unsupervised learning to detect malicious activity in its network traffic based on the patterns of known malicious activity. This would allow the company to protect its network from attack and to prevent data breaches.

Reinforcement Learning

Reinforcement learning (RL) is a type of machine learning where an AI agent learns to behave in an environment by trial and error. The AI agent is not explicitly programmed with rules or instructions, but instead learns by interacting with the environment and receiving rewards for taking actions that lead to desired outcomes.

Reinforcement learning has been used for a variety of tasks, including:

• Game playing: RL has been used to train AI agents to play games such as Go, Chess, and StarCraft. These AI agents are able to compete at a superhuman level, and have even defeated human champions.
• Robotics: RL has been used to train robots to perform tasks such as object manipulation and navigation. These robots are able to learn from their mistakes and improve their performance over time.
• Finance: RL has been used to develop trading algorithms that can automatically trade stocks and other financial instruments. These algorithms are able to learn from market data and make profitable trades.

Deep Learning

Deep learning is an advanced type of machine learning that is better suited for complex tasks involving unstructured data, such as image recognition, natural language processing, and machine translation. It is a powerful tool that has the potential to revolutionise the way we interact with computers.

As a way of explaining deep learning, imagine you’re trying to teach a child how to recognise a cat. You’d probably show them several pictures of cats, pointing out the common features like whiskers, a tail, and pointy ears. Over time, the child would start to understand what makes a cat a cat and could identify one on their own.

Deep learning works in a similar way, but with computers instead of children. You feed the computer a lot of data – for example, thousands of pictures of cats. The computer analyses this data and learns to recognise patterns that define what a cat looks like. Once it’s learned enough, you can show it a new picture it’s never seen before, and it can tell you whether there’s a cat in it.

Generative AI: Creating New Content

If there’s one thing that would explain why AI became such a hot topic in 2022/2023 it is probably the advent of Generative AI.

Generative AI is a type of artificial intelligence that can create new data, such as text, images, or music. It does this by learning from existing data and then using that knowledge to generate new content that is similar to the data it has learned from.

The power of Generative AI lies in its ability to go beyond simple replication, and produce content that is original and creative. It’s like having an AI-powered artist or writer at your disposal, capable of generating new ideas and creations, based on the data it has learned from.

However, while Generative AI has made significant progress, it is still a rapidly evolving field. With each iteration, it continues to learn from past experiences and gains access to new data, enabling it to improve its performance and generate even more realistic and informative content.

As Generative AI continues to advance, we can expect even more exciting applications in various fields such as art, literature, music, and more. The possibilities are endless as this technology unlocks new ways for machines to create and innovate alongside humans.

Generative AI learns the patterns, structures, and features in the data it is trained on, then generates new content based on this knowledge.

The launch of ChatGPT

In late 2022, OpenAI launched its ChatGPT chatbot for public use. This was one of the first occasions that the general public had access to a Generative AI tool. The chatbot was able to generate realistic and informative text, and it was also able to answer questions and provide information.

People were excited about the potential of this technology, and they were eager to see what it could do. This led to a surge in research and development in AI, and it also led to the development of thousands of new AI-powered products and services. Hardly a day goes by now when there isn’t an announcement of some new AI tool.

The release of ChatGPT was a major event in the history of AI. It was the first time that the general public got to see Generative AI in action and to begin to understand the power of AI.

Here are some of the reasons why ChatGPT caused an explosion of interest in AI:

• • It could create text that looked just like what humans write, such as articles, web pages, e-books, email text, poems, song lyrics etc. etc. This was a significant revelation because it showed that AI could make text that people couldn’t tell apart from what other people write.
  • It was able to answer questions and provide information in a more useful way than search engines. This made it a valuable tool for research and learning.
  • It was easy to use. Anyone could use ChatGPT, even if they didn’t have any experience with AI.

Generative AI in business

Generative AI holds massive potential for businesses both large and small. Just in the field of marketing alone, we are already seeing organisations use tools like ChatGPT to grow their businesses. Here are some examples:

• • Generating leads: Generative AI can be used to generate leads by creating realistic and engaging content that attracts potential customers.
  • Optimising website content: Generative AI can be used to optimise website content by creating headlines, images, and other content that is more likely to attract visitors and convert them into customers.
  • Creating personalised content: Generative AI can be used to create personalised content for each customer. This can include things like product recommendations, email marketing campaigns, and social media posts.
  • Improving customer service: Generative AI can be used to improve customer service by creating chatbots that can answer customer questions and resolve issues.

Peering into the future of Generative AI

Since the launch of ChatGPT we’ve seen many more Generative AI platforms launch. These include:

• • Google Bard: A competitor of ChatGPT that can generate creative content, translate languages, and answer complex questions.
  • Copilot: Microsoft’s AI assistant that is embedded in Microsoft 365 apps.
  • Midjourney: A text-to-image Generative AI platform. Midjourney can create stunning images based on a text description you give it.
  • DeepMusic: This tool can create music that is indistinguishable from human-composed music.
  • Creatopy: A platform that can be used to create custom marketing materials that perfectly match a brand’s identity.

These are just a few of the thousands of Generative AI tools and platforms that have sprung up since the launch of ChatGPT.

Generative AI is a rapidly developing field, and it is difficult to predict exactly what the future holds. However, one thing is certain. Generative AI models will become increasingly realistic. This is due to the use of larger datasets and more powerful algorithms. In the future, we can expect generative AI models to create images, videos, and text that is indistinguishable from real-world content.

The Dangers of AI

Artificial intelligence, as we’ve seen, has the potential to be a powerful tool for good. However, it is important to acknowledge the potential dangers and negative side effects associated with the technology.

Here are some of the dangers and negative aspects of AI:

• • Job displacement: As mentioned previously, AI is already being used to automate tasks that were once done by humans. As AI technology continues to develop, it is likely that even more jobs will be automated. While it is true that new jobs related to AI will be created, the rate at which jobs are being displaced may outpace the creation of new ones, resulting in a significant impact on employment. This could potentially lead to economic inequality and social upheaval if not addressed adequately.
  • Bias: A significant danger of AI lies in its ability to perpetuate biases and discrimination. AI systems are trained on data that is collected from the real world. This data can be biased, and this bias can be reflected in AI systems, leading to unfair treatment in areas like hiring practices, loan approvals, and criminal justice systems, further exacerbating existing inequalities.
  • Loss of privacy: Privacy and security are also major concerns when it comes to AI. As AI technology becomes more sophisticated, it has the potential to gather and analyse vast amounts of personal data. This data could be used to track people’s movements, monitor their activities, and even predict their future behaviour.
  • Fake news: Fake news is not a new phenomenon but AI has the potential to amplify its impact significantly. AI can generate convincing text, images, and even videos (known as deepfakes) that can be almost indistinguishable from real ones. This makes it easier for malicious parties to create and disseminate false information on a large scale. Through social media platforms, fake news can spread rapidly, leading to misinformation and confusion among the public. This can have serious consequences, such as influencing elections, inciting violence, or damaging reputations.
  • Ethical dilemmas: AI-powered autonomous systems pose risks as well. For instance, self-driving cars have the potential to significantly reduce accidents caused by human error. However, they also raise ethical dilemmas when faced with decisions that involve prioritising the safety of different individuals in potentially life-threatening situations.
  • Autonomous Weapons: AI can be used to develop autonomous weapons systems that could kill without human intervention. These weapons systems could pose a serious threat to global security.
  • Superintelligence: Some AI researchers, and well-known tech personalities, have raised concerns about the possibility of creating an AI that surpasses human intelligence, known as superintelligence. This raises existential questions about the potential loss of control over technology and its implications for our society.

“AI is a rare case where I think we need to be proactive in regulation instead of reactive. Because I think by the time we are reactive in AI regulation, it’s too late.” – Elon Musk

While AI offers tremendous opportunities, it is crucial to also recognise the negative aspects of it. To ensure that AI benefits society as a whole, while minimising potential harm, it is paramount that safeguards are established to prevent any unintended harm or misuse.

Governments around the world have recognised these concerns, and have started introducing regulations around AI. These regulations aim to address issues such as data privacy, algorithmic transparency, accountability, and ethical considerations. It remains to be seen whether these new regulations can ensure that AI is only used for good.

Why you need to learn about AI

AI cannot be ignored. Whether we like it or not, It will impact each and every one of us, in both our personal and professional lives.

As already mentioned, AI will displace many jobs, as well as create some new ones. If you have a desk job, for example, then you can expect to see AI make rapid inroads into your profession. So if you’re not using ChatGPT, or an equivalent tool, to help you be more productive in your workplace, then you are already slipping behind. Employers will increasingly look to hire people who are familiar with using AI tools. If you want to protect your future career prospects it will therefore be essential to gain AI skills.

As a business leader, it will border on professional negligence to ignore AI in your company. Regardless of the size of your company, or the industry you operate in, you should be investigating AI and learning how it can help your business. Whether it’s in operations, marketing, customer service, data analytics, or other areas of the business, there is a role for AI in every company. Any business choosing to ignore AI is running a very real risk that they will get overtaken by competitors who invest in using AI. (For more on this see my blog post “Why Ignoring AI Could Be Detrimental for SMEs“.)

Conclusion

As I write this, it is late July 2023, 8 months since the release of ChatGPT. The hype we’ve experienced around AI in that short time is almost unprecedented. In my many years in IT, I can only think of the hype around the internet in the late 90s as being in any way comparable.

Having lived through the emergence of PCs, the internet revolution, the advent of cloud computing and the rise of mobile technology, I can honestly say, without any hint of hyperbole, that I think AI is the most significant technological advancement in my lifetime. Everything that has happened over the last 30+ years has laid the foundation for AI to build upon, and I believe that we have reached a significant inflection point in our use of technology.

It’s still early days in the history of AI, but its potential seems limitless. AI will be transformative for our world, in ways we can’t even imagine today. Predicting what will happen with AI is almost impossible, but what we can say for certain is that it will get better with each iteration as it learns from more data, and that it will become pervasive in all aspects of our lives.

However, as we embrace the potential of AI, it is essential that we as humans do so responsibly. This means addressing concerns around fake news and deepfakes, bias in algorithms, ensuring transparency in decision-making processes, and safeguarding privacy and security. It also requires us to  consider the ethical implications of AI, such as its impact on employment and social inequality.

Disclosure

I’ve been writing articles on technology for nearly 30 years. I’m also a published author. So when it came to creating this post, I was quite comfortable writing this under my own steam. However, given the topic of the article and the emphasis on Generative AI, it would have been remiss of me not to use AI to help me.

I would classify this article as an AI-assisted post. I could have let AI generate the whole article in a few minutes, but I wanted to ensure that it had my stamp on it and reflected my knowledge of AI. I also wanted to ensure that it read in the way I would have written it, if I had created it without AI help. Of course, I could have trained AI in my style, but I actually like writing, so I didn’t go down that route for this post. 

So the article is definitely mine, and has my imprint on it, but there are also sections that AI helped me with and made my life easier. However, even where I’ve used AI generated content, I’ve still edited it and added my own contribution, mainly because I’m so used to writing and I couldn’t resist making some modifications to enhance the readability of the post.

The most important point is that the article reflects my knowledge gained from 40+ years in the IT sector, and my immersion in all things AI in the last 12 months. I’ve been predicting the dominance of AI for some time, and with the emergence of ChatGPT and other tools/platforms, we are now at the beginning of a new technological revolution.