Generative AI, by Midjourney
Generative AI, by Midjourney

I have been poking fun at chatbots recently, but in this article I wanted to be clear: Generative AI is a game changer, and you need to start preparing right now. I have already alluded to this topic in my previous posts, but I wanted to tackle this specific issue head on in more detail.

In this article I will explain:

  • What generative AI is
  • Why it is different from anything we have seen so far
  • Why it will have an enormous impact on small businesses, and
  • What you can do about it right now

What is generative AI?

Generative AI refers to artificial intelligence techniques that are capable of generating new data or content that mimics human creations. This includes generating text, images, videos, and even music or other forms of creative expression.

Here is an example:

Write a humerous limerick to apologise to my wife for
forgetting to purchase milk on my way home from work.
There once was a man in a hurry,
Whose memory could be quite blurry.
He left work with a grin,
But forgot milk for his kin,
Now his wife's angry, oh dear, how scurry!

But fear not, my dear forgetful friend,
Your apology can make amends.
Say sorry with a smile,
And go the extra mile,
To make sure this mistake won't happen again!
Rewrite in the style of a shakespearean sonnet.
Oh dearest wife, forgive this foolish man,
Whose mind oft wanders and forgets his duty.
His heart doth ache with guilt, his cheeks are wan,
For milk he did not bring, that vital beauty.

The market on his path did call his name,
And he did think of other things beside.
But now his heart is filled with burning shame,
And tears of penitence he cannot hide.

So let me make it up to thee, my dear,
With chocolates sweet and roses bright as morn.
And I shall pledge to never more appear,
Forgetful, careless, or so deeply forlorn.

My love for thee is boundless, as the sea,
And in my heart, I pray thou'lt pardon me.

Surely not useful in a professional setting, but very impressive!

One of the most well-known examples of generative AI is OpenAI's GPT (Generative Pre-trained Transformer) series of models, which are capable of generating human-like text based on a given prompt. Another example is the StyleGAN system, which can generate photorealistic images of people, animals, and other objects.

Why is generative AI different?

Other than generative AI, some of the major areas of study and application of AI technologies are pattern recognition, categorization, decision support, and natural language processing.

Pattern recognition is an application of AI that identifies and analyzes patterns in data, images, or other types of input. Some examples of pattern recognition systems include facial recognition systems used for security or identifying individuals, speech recognition systems used in virtual assistants, voice-controlled devices, or transcription services, and object recognition systems used in autonomous vehicles, robotics, or medical imaging. These systems work by analyzing input data and identifying patterns, which can then be used to make predictions, classifications, or other types of decisions.

Categorization is an application AI that classifies or categorizes data based on specific criteria. Some examples of categorization systems include recommendation systems used by e-commerce sites, streaming services, or social media platforms, spam filters used in email or messaging applications, and fraud detection systems used in financial institutions or e-commerce platforms. These systems work by analyzing data and assigning it to specific categories based on predefined rules or machine learning algorithms.

Decision support is an application of AI that provides assistance in decision-making processes. Some examples of decision support systems include predictive analytics used in forecasting, sales, or risk management, supply chain optimization systems used in logistics or manufacturing, and medical diagnosis support systems used by healthcare professionals or patients. These systems work by analyzing data, identifying patterns or trends, and providing insights or recommendations that can be used to inform decision-making.

Natural language processing (NLP) is an application of AI that focuses on understanding and processing human language. Some examples of NLP systems include language translation services such as Google Translate, sentiment analysis used by companies to monitor brand perception on social media, and chatbots used for customer service, technical support, or information retrieval. These systems work by analyzing language data and using machine learning algorithms to extract meaning, classify sentiment, or generate responses in natural language.

Each of these application types have been around already for a number of years.

The main difference between these applications and generative AI is that the applications above are used to analyze and interpret existing data. In order to work, they require large quantities of data as input. Generative AI creates new content. Although it requires a huge language model to work, generation of new content can be done with a very simple prompt.

I would argue that analyzing and predicting are functions that we humans already expect computers to do. Although it is certainly an impressive technological feat, it is no longer something that amazes us; it has become such an integrated part of our daily lives that we take it for granted and don't even notice it anymore.

Creation, however, is something that we have typically associated with being human. Seeing computers create similarly to humans is something that is shocking: inspiring and exciting for some, fear instilling in others. It is something that gives us more of an emotional reaction.

Personally, I react rationally and logically to a recommendation engine or a translation engine, just as I would react to calculating 2+2 on my calculator (because I am no longer able to do simple arithmetic in my head 🤪). However, I can get quite an emotional reaction when interacting with technology like ChatGPT. Although I know better, I find that I often thank it, praise it, and wish it good night.

Thank you. That's all for today. Nice work! Good night.
You're welcome! It was my pleasure to assist you. Have a good night!

Obviously, AI is not sentient, and likely will never be so. But that is a topic for philosophers and novelists to ponder. The fact is, it doesn't matter. If AI can do things better than us humans can do, then a portion of society will be able to profit from it. If there are people who can profit, then it will be promoted and leveraged heavily. It will undoubtedly change the business landscape.

How will generative AI impact small businesses?

Although I generally try to avoid making futuristic predictions, I think that it is not unreasonable to predict that once integrated fully in to our working environments, generative AI will be able to relieve us from most language-intensive tasks. It will undoubtedly change the nature of work, hopefully for the better.

Language-related work will require less energy and focus, making it easier for us to be creative and productive. This could be from anything to communicating mundane things, to drafting important legal documents.

As I discuss in a previous article, it is conceivable that intensive STEM training could become a thing of the past, as logical thinking will be replaced by more "natural" thinking, with people interfacing with machines by using natural language, instead of programming.

If so, the nature of work will shift from the rational to the emotional. Since AI will do the heavy lifting for us, and since we will interact with AI by using natural language, our thinking will be less logical and methodical, and more intuitive and emotional.

Our interactions with technology will become increasingly verbal. This means: no more keyboards. Personally, I will be shedding no tears over this one.

There are surely many other predications one can make, but I will stop here.

One thing of which I am certain is that technology vendors will be falling over each other to integrate AI into everything. The race is on. Unlike other fads, like blockchain, which are indeed interesting and useful technologies, AI is on another level because it affects us so viscerally. The impact will be large, and the impact will be felt quickly.

What should you do about it?

Companies that don't leverage AI will be left behind. However, with all the noise and hype, what is one to do?

You should immediately take another look at your value proposition. If it relies on something that generative AI can do, then you need to take immediate action, because the shift will be quick, and it will be great. If you do not start making immediate plans to update your business model, you will be blindsided.

If generative AI cannot fundamentally replace or significantly enhance your value proposition, then you have the luxury of time. You will still have to integrate it to make incremental changes, but your business is not in immediate danger. More likely than not, it will be more difficult not to integrate AI, as all providers will be rushing to integrate it into their services, whether you want them to or not.

  1. Re-evaluate your value proposition
    • Is it still valid, or has your business shifted over time, so your value proposition needs updating?
    • Is it clear, or does it need another iteration of refinement?
  2. Determine if your value proposition relies on work that can be significantly replaced by generative AI
    • Are you a content creator?
    • Are you in any other category that is deemed "at risk"?
  3. If your business model is impacted, determine very clearly which parts are impacted, and which are not
    • Make sure you create a plan to invest all-out in AI for those areas that will be most impacted
    • Start looking more closely at those areas that cannot be replaced by AI, as this will be your competitive advantage

So much is uncertain and there are surely many surprises to come, but it is possible to start preparing for change.