Introduction
Artificial intelligence (AI) has been around for some time, although it may not have been as obvious as it is today. Already in the 1980s, we could see examples of AI in games such as Pac-Man, where the yellow figures were controlled by a simplified version of AI. But it wasn't until the 1990s, when companies like Google started implementing AI to analyze user behavior and improve their services, that AI began to be used more widely. Today, AI has become an integral part of our digital life. Our mobiles are absolutely the clearest example where our cameras use AI to calculate where the focus should be for the image to be sharp. Social platforms like TikTok and Facebook rely on AI to control, adapt and display content according to user preferences and usage patterns.
The image was produced with the help of Microsoft Bing's AI.
The trend of AI in the world
If we do a simple investigation of how the search trend on Google looks like in the world after the word AI, we see that the last 12 months have been steadily increasing. One guess is that the trend will continue and that it is high time as entrepreneurs to start absorbing information about what AI is and how it can facilitate and make your business more efficient.
What is Generative AI?
One of the most exciting advances in AI today is the development of generative AI, which has exploded over the past year. Generative AI is about giving an instruction, which is usually called a "prompt", to an AI model and getting a unique response. The technology has been made possible through three components; 1) Cloud-based computing capacity – E.g. faster and more accessible cloud services 2) Large amounts of data that AI can train on 3) Advanced algorithms that drive progress in generative AI. It is also important to mention that generative AI has been around for a long time, but it is only in the last year that services such as chatGPT and Bard AI have become available which make the user-friendliness and interface inviting for "ordinary users".
The importance of prompting
An important factor in generative AI is the quality of the instruction or prompt given to the AI model. The more detailed and personalized the instruction, the better the result. It is important to describe in detail and clearly what you want an answer to in order for the AI model to succeed in generating a unique and reliable result. For those who want to practice prompting and understand how instructions affect performance, the site is recommended learnprompting.org.
What is ChatGPT?
A well-known model today in generative AI is ChatGPT (Chat Generative Pre-trained Transformer). It is an AI model that has been trained on large amounts of anonymous text data and can generate authentic and natural language. Using ChatGPT, AI systems can have interactive conversations and create texts that resemble human writing. The model has revolutionized the way we interact with AI and opened up new possibilities in areas such as customer service, support and generation of text content.
An important part of this is also what the internet foundation mentions in its basic course in AI (which is recommended to anyone who wants to understand AI from the ground up).
“ChatGPT is a language model. It predicts words and is good at imitating the shape of human writing. But one should not take what is written for facts. It imitates human texts, but has no built-in capacity for source criticism or truth.”
The image was produced with the help of Microsoft Bing's AI.
AI and e-commerce
According to the Pricerunner report “E-retailer index for Q1 2023” 21 percent of the surveyed e-commerce stores have already implemented AI in their solutions. This indicates that AI is becoming increasingly popular and that companies that use this technology can have a significant advantage over their competitors that do not explore and implement AI. AI offers a range of possibilities for e-retailers, from e.g. automating the creation of product descriptions to analyzing data from different systems, improving marketing strategies and streamlining customer service.
A clear example from a company that uses AI is Athea. They have developed an AI employee at their customer service named Frank. Frank helps, manages and distributes the cases to the right colleague. Today, Frank can handle 45% all by himself out of the 900 incoming e-mails the company receives daily.
Another example is NA-KD which uses a service to automatically generate and translate product texts into different languages with the help of AI. Such an implementation can reduce the cost by approx. 70%. Therefore, it is important as a company to find what can be optimized without losing quality and start pilot projects to find these keys.
Become an AI company
- Start pilot projects that can make your business more efficient today.
- Create an in-house AI team – Involve colleagues.
- Train your entire company around AI. There are good courses online, but it is also possible to invite AI experts who can tell you about the advantages, disadvantages, risks and what business opportunities exist.
- When you have labled, tested and understood how AI can streamline and optimize your operations and business, it is time to set an AI strategy in general for the company.
- Create an internal and external communication plan
Summary
AI has already become an integral part of e-commerce and offers a range of opportunities for companies to automate processes, understand customers better and create more engaging experiences. Those who embrace AI technology and exploit its potential are likely to have a competitive advantage in the rapidly changing e-commerce industry. Invest in starting pilot projects internally and try to understand as quickly as possible how AI can streamline and optimize your business. Both financially but also procedurally.
Do you have an idea or want to better understand what AI is and what it can do for your particular company. Do not hesitate to contact us at eKompass.
English (UK) 
Svenska