AI in Retail Part 1: What is AI and How Will It Impact Retail in the 2020s?Retail
Since the release of OpenAI’s ChatGPT in late 2022, artificial intelligence (AI) has become one of the hottest topics for businesses of all types. The hype is real, and for good reason: new generative AI products are a leap forward in accessibility and practical application of AI technology for everyday use.
What do advancements in AI mean for retail? What is AI, how can it be applied in a retail setting, and how will it impact retailers? What are the opportunities and risks it presents for retail organizations?
These are the questions we seek to explore and answer in our new 3-part series on AI in Retail, partly based on IHL Group’s original new Agilence-sponsored report, Retail’s AI Revolution.
What are the 3 main types of AI?
Artificial intelligence (AI) is a broad concept with a broad definition. According to Wikipedia, pulling from Patrick Henry Winston of MIT’s 1984 academic book Artificial Intelligence, AI is “intelligence–perceiving, synthesizing, and inferring information–demonstrated by machines, as opposed to intelligence displayed by humans or other animals.”
Broadly, AI refers to the development of computer systems that can perform tasks which typically require human intelligence, such as recognizing speech, interpreting data, and making decisions. AI systems are designed to learn from experience, adjust to new inputs, and perform human-like tasks with great accuracy and speed.
AI involves many subfields, including machine learning, natural language processing, generative AI, computer vision, robotics, and neural networks. For our purposes, based on IHL’s methodology, we’re going to break out AI into three main types with potential to impact retailers: Machine learning, generative AI, and artificial general intelligence (AGI).
Machine learning is a subset of AI that involves training algorithms to learn patterns in data without being explicitly programmed. As the algorithms “learn,” they can make predictions or decisions based on that data. This allows AI systems to improve their performance over time as they encounter more data, becoming more accurate and efficient at the tasks they are designed to perform.
Machine learning algorithms are already used in a wide variety of applications, such as email filtering; fraud detection; image and speech recognition; computer vision and robotics; medicine; the recommendation systems that power your experience in Netflix, Amazon, YouTube, and social media platforms; and many more. Because these algorithms rely on the data they’re fed, data hygiene is extremely important to benefitting from machine learning.
How AI/ML Will Impact Retail
AI/ML is already having a significant impact on the retail industry, and it has the potential to transform many aspects of retail in the future. Some emerging uses cases of AI/ML in retail include:
- Personalization: AI/ML algorithms can analyze customer data to provide personalized recommendations and experiences, such as targeted offers, customized product recommendations, and personalized marketing messages.
- Inventory management: AI/ML algorithms can help retailers optimize their inventory by predicting demand, identifying slow-moving products, and automating restocking processes.
- Pricing optimization: AI/ML algorithms can analyze market data and consumer behavior to optimize pricing strategies, predicting which products will sell and when to offer discounts or promotions.
- Customer service: AI/ML-powered chatbots and virtual assistants can provide 24/7 customer service, answering questions, resolving issues, and providing customer support without the need for human intervention.
As AI/ML develops in the future, it’s likely to unlock new possibilities such as autonomous stores that use sensors, cameras, and other technologies to track customers, products, and inventory (such as those piloted by Amazon Go), more advanced supply chain management capabilities, and better predictive analytics. However, these are still a way off from realization.
Implementing AI/ML technologies can be challenging and require extensive data cleansing and tagging, which we’ll dive more into later in this series.
#2. Generative AI
Generative AI is a broad term that encompasses any type of AI that creates new data (often in the form of images, audio, or text that is similar but not identical to its training data). Generative AI uses neural networks to create new data based on patterns found in existing data. Unlike machine learning algorithms, which are typically used for classification or prediction tasks, generative AI is used for creative tasks such as image and music generation.
Generative AI is a subset of machine learning, and generative AI systems are created by applying machine learning models. The main difference between machine learning and generative AI is their purpose. Machine learning focuses on predicting outcomes or classifying data based on patterns found in existing data, while generative AI focuses on creating new data based on those patterns. Both types of AI have unique applications and capabilities, and they often work together in AI systems to accomplish complex tasks.
Types of Generative AI Models
The most well-known generative models are Large Language Models (LLMs), which are used by services such as ChatGPT to generate text by probabilistically predicting the character sets which should follow a given natural language prompt. Large language models work by using neural networks to analyze and understand the structure of language. The model is trained on a large corpus of text data, and then uses that knowledge to generate new text based on a set of input prompts. For example, a large language model could be trained on a dataset of news articles, and then be used to generate new articles based on a given topic or headline.
Other types of generative models include Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), which are used to generate images and other types of data. With GANs, two neural networks are pitted against each other: one network generates new data, and the other network evaluates that data to determine whether it is real or fake. Over time, the generative network learns to create more realistic data, while the evaluating network becomes better at distinguishing between real and fake data. GANs can be used to create photorealistic images, videos, and even three-dimensional models.
The most common applications of Generative AI include:
- Image and video generation: Generative AI can be used to create new images or videos that look like they were created by humans.
- Music and sound generation: Generative AI can be used to create new music or sound effects that are similar to existing music or sound effects.
- Text generation: Generative AI can be used to generate new text, such as news articles, product descriptions, or even entire books.
How Generative AI Will Impact Retail
For retailers, Generative AI can be used to create unique product designs, packaging, or even store layouts. It has incredible potential to disrupt the knowledge worker and impact the Sales as well as the General Administrative part of retail, greatly increasing productivity. Here are some examples of how generative AI could impact retail:
- Custom product design: Generative AI can be used to create custom designs for products such as clothing or furniture. By inputting customer preferences and specifications, the algorithm can generate unique designs that meet the customer's needs.
- Virtual try-ons: Generative AI can be used to create virtual try-on experiences that allow customers to see how products such as clothing or makeup will look on them before making a purchase.
- Chatbots: Generative AI can power chatbots that provide personalized customer service, answering questions and aiding with purchasing decisions.
- Content creation: Generative AI can be used to create product descriptions, social media posts, and other content that is tailored to the customer's interests and preferences.
Artificial General Intelligence (AGI)
Artificial General Intelligence (AGI) is a hypothetical form of AI that can perform any intellectual task that a human can. AGI is sometimes referred to as "strong AI" or "human-level AI," and it represents the ultimate goal of AI research. Unlike the types of “weak AI” mentioned above, strong AI does not rely on specific programmed models for performing narrow tasks. A true AGI system would be able to learn, reason, and solve problems across a wide range of domains and contexts, transfer knowledge and skills from one domain to another, and be able to adapt to new situations and environments.
AGI is a common topic in science fiction and when researchers express concern over AI as a threat to humanity, they’re talking about AGI. The timeline for AGI development is a subject of debate among experts; some believe it is years or decades away, some believe it may take a century or longer, while others believe it will never be achieved.
Retailers don’t need to worry about AGI just yet, as it’s largely theoretical, with optimistic forecasts putting it at 3-5 years away. However, if achieved, AGI could potentially transform the retail industry by automating complex tasks, such as product design and supply chain management, and providing a level of customer service that is indistinguishable from a human representative.
The realm of AI is saturated with a mix of genuine breakthroughs and exaggerated claims aiming to capture attention, but akin to electricity and the internet, AI holds transformational power that can be harnessed for positive or negative outcomes. However, one undeniable truth applies to both organizations and individuals: while AI may not directly replace jobs, those who possess expertise in leveraging AI tools will have a distinct advantage over those who do not.
How will AI economically impact retail?
In the Agilence-sponsored report by IHL Group, “Retail’s AI Revolution,” IHL’s analysts adopt a comprehensive approach to assess the overall economic influence that each AI type will have through 2029.
Using their Worldview IT Spend Model as a foundation and refining the data based on 10 Line-of-Business Categories (including BI, eCommerce, Sales and Marketing, Distribution and Supply Chain, etc.), IHL calculated the projected impact of the three types of AI mentioned above, broken down by three categories:
- Impact by tier (smaller retailers vs. larger chains)
- Impact by retail segment (such as Food/Grocery, Pure Play eCommerce, Mass Merchants/Hypermarkets), Specialty Hard Goods)
- Impact by region (North America, Asia, EMEA, LATAM)
Worldwide, IHL is forecasting the overall economic impact through 2029 to be $9.2 trillion. To read the complete findings, including the impact by the three categories above, download the report for free today.
In Part 2 of this series, we’ll explore the opportunities AI presents for retailers, including:
- Sales & Revenue Growth Opportunities
- Product Cost/Supply Chain/Gross Margins Impact
- Reducing SG&A with Generative AI
3 Key Opportunities for Retailers to Improve Margins with AIIn 2023, organizations of all kinds are grappling with the breakthrough products (and incessant hype) of Generative AI. For r...
AI Risks and How to Start Now with AI – AI in Retail Part 3Since the release of OpenAI’s ChatGPT at the end of 2022, there’s been a big question on the mind of retail leaders: how will...
Webinar Recap - Retail's Digital Journey: How Winning Retailers Are Controlling Shrink and ThrivingIn our recent webinar, Retail’s Digital Journey: How Winning Retailers are Controlling Shrink and Thriving, Greg Buzek of IHL...
Subscribe to our blog
Receive free educational resources like exclusive reports, webinars, and industry thought leadership articles straight to your inbox.