2023 Global Generative AI Landscape

Generative AI (or “Gen AI”) technology is on the brink of transforming the retail industry, presenting unprecedented opportunities for businesses to enhance their productivity and competitiveness. According to an article recently published by McKinsey, Gen AI has the potential to unlock significant economic benefits across various sectors, with the retail industry standing to gain immensely from this technology.

Defining Gen AI

In case you are still unclear on the precise definition of the term Gen AI, it is a type of AI that involves training algorithms to understand and replicate patterns found in data, thus enabling machines to generate new content that closely resembles human-created content. By leveraging Gen AI, retailers can streamline operations, improve customer experiences, and ultimately drive growth.

How Gen AI will Transform Retail

Per the McKinsey report alluded to above, one of the primary advantages of Gen AI lies in its ability to augment human capabilities. By automating repetitive and time-consuming tasks such as inventory management and demand forecasting, retailers can free up human resources to focus on higher-value activities. This not only increases operational efficiency but also empowers employees to make more strategic and creative contributions to the business.

Gen AI can also be leveraged to enhance customer experiences in the retail industry. By analyzing vast amounts of data, including customer preferences, purchase history, and social media interactions, AI algorithms can generate personalized recommendations and targeted advertisements. This level of customization creates a more engaging and tailored shopping experience, leading to increased customer satisfaction and loyalty.

Moreover, Gen AI has the potential to revolutionize product development and design. By analyzing market trends and consumer preferences, AI algorithms can generate novel ideas, prototypes, and designs, thus reducing the time and cost associated with traditional content and product development cycles. This accelerated innovation process enables retailers to bring new marketing campaigns and products to market more rapidly and efficiently, gaining a competitive edge in an environment where consumer preferences and demands are changing at an accelerating pace.

McKinsey also emphasizes the impact of Gen AI on supply chain management. By optimizing inventory levels, predicting demand patterns, and automating procurement processes, retailers can minimize stockouts and reduce wastage, leading to significant cost savings. Gen AI can also enhance supply chain transparency, enabling retailers to track and trace products more effectively, ensuring compliance with quality standards and regulatory requirements.

History of Gen AI

It should be noted that Gen AI technology is not brand new. The technology was introduced in the 1960s via chatbots. But it was not until 2014, with the advent of generative adversarial networks or GANs (a type of machine learning algorithms), that Gen AI could create remarkably realistic images, videos, and audio of real people. The recent buzz around Gen AI has largely been driven by the simplicity of new user interfaces for creating high-quality text, graphics, and videos in a matter of seconds. Chat GPT — a popular Gen AI platform — set a record for the fastest user growth in January when it reached 100 million active users two months after its launch. The technology has continued to improve since then, with it now hitting an inflection point where it has piqued the interest of everyday people and businesses.

Challenges of Gen AI

There are various challenges that come with adopting Gen AI technology. Retailers need to invest in data infrastructure, talent acquisition, and employee upskilling to effectively harness its power. Additionally, ethical considerations surrounding data privacy, bias, and accountability must be addressed to ensure responsible and trustworthy use of any AI driven technology.

There is still a long way to go on these fronts, and retailers would benefit from taking a prudent approach during these early innings of the technology’s adoption within the enterprise.

Gen AI for Retail: Global Market Map

To help retailers and brands think through how they can adopt Gen AI in their respective businesses, we put together a startup map — with companies organized by the nature of the capabilities they can offer to customers. Companies highlighted on the map are represented under three prioritized strategic areas, and are further categorized based on their primary use case.

Connecting with these companies is a great way to get acquainted with the latest advancements in the field of Gen AI and experimenting with all the existing capabilities. The following is a summary of the categories and subcategories under which the startups have been organized:

Product and Merchandising

In this category, companies are embracing human-machine co-creation to redefine product design processes. Given Gen AI’s core ability to produce new content, product design is a clear and significant — and thus prioritized — use case for the technology.

Product Ideation: Design and Visualization

This subcategory features startups that enhance the creative process by prototyping collections or generating new designs based on specified parameters, such as colour palette, fabric, and trim, to facilitate product ideation. Gen AI-powered design visualization and conceptualization tools allow designers to efficiently iterate and refine their concepts, saving resources and accelerating time to market. As an example, fashion design and photoshoot studio Resleeve can support experimentation by generating new designs or numerous variations of existing concepts using simple text prompts, and expedite the design process by converting a designer’s sketches into high-fidelity mock-ups.

While excitement for this technology is warranted, it is important to embrace these tools with caution. Questions concerning intellectual property rights and whether one can claim ownership of AI-generated designs are being actively debated, especially since models trained on publicly available data without sufficient safeguards against copyright violations can imitate existing works, generate derivative designs, and consequently compromise brand integrity.

Considering the current state of play, these applications can still be integrated into existing workflows if they’re used to support early-stages of ideation and mood boarding. Then, designers can add original elements and company branding to the mix.

Marketing

Gen AI is poised to revolutionize the entire marketing function by streamlining content production processes and enhancing productivity. Given the technology’s ability to produce hyper-personalized text and visual content at scale, the “Content Creation and Advertising” and “Product and Marketing Copy” subcategories represent some of the most promising use cases that align best with Gen AI’s core capabilities.

Content Creation and Advertising

This subcategory consists of content writing and video / image generation tools that accelerate campaign development, thus allowing retailers and brands to test and iterate content based on what’s resonating with their community more easily. The technology’s ability to recognize patterns in viral content and produce imagery with varying specifications can be harnessed to create compelling campaigns tailored to the individual consumer, which can help optimize ad spend and save valuable time in the process.

A word of caution: large language models trained on publicly available data can replicate content produced by other retailers and brands, and pose branding recognition risks. Using the technology in the early planning stages of content development to facilitate ideation can help marketers to mitigate this risk and ensure that content stays true to the brand’s unique identity and heritage.

Product and Marketing Copy

This subcategory consists of AI-powered writing assistants as well as copywriting software that auto-generate marketing copy or SEO-optimized product descriptions. The utility of Gen AI in copywriting goes beyond expediting the writing process; a number of platforms highlighted in this subcategory can optimize copy for better performance and conversions by leveraging customer insights and historically successful approaches. Moreover, AI writing assistants can provide guidance throughout the copywriting process by suggesting relevant keywords for SEO and providing real-time feedback to ensure consistency and alignment with the brand’s voice.

Introducing Gen AI to copywriting requires careful consideration, however, since large language models can make factual errors and produce inaccurate information. At this stage, AI-generated copy needs to be manually reviewed and companies need to decide how much human oversight is appropriate.

Trend Analytics and Consumer Insights

The startups in this subcategory harness Gen AI’s advanced natural language processing capabilities to analyze customer feedback and extract valuable insights for product and CX teams. Thus, these tools enable retailers to fine-tune their offerings to optimize spend and elevate customer experiences.

Digital Commerce and Consumer Experience

Synthetic Models and Product Photography

Startups belonging to this subcategory streamline content creation for product listings by generating product photography tailored to the retailer or brand, the product’s specific applications, and/or by generating a diverse range of synthetic fashion models to showcase products on. With the ever-increasing speed of modern trend cycles and evolving consumer tastes, these tools allow retailers and brands to shorten content production timelines, lower operational costs, and continually adapt product imagery. AI-generated fashion models can additionally be customized based on skin tone, body type, size, and age to reflect a diverse customer base. Veesual for example offers a product that allows site visitors to view garments on a model of their choosing — a model they identify with, for example — to foster more personal and engaging shopping experiences.

The advent of AI-generated models has been a controversial topic, however, and early adopters of the technology have faced backlash (see here). Critics argue that AI models may displace human models, particularly models of colour, or perpetuate unattainable beauty standards.

Thus, retailers and brands looking to explore this approach need to consider how it will resonate with customers and may want to include disclaimers on product imagery generated with synthetic models.

Virtual Try-on Tech

Startups in this subcategory are creating applications that enable customized virtual apparel try-on experiences to inspire customer loyalty and reduce the likelihood of returns. Using Gen AI, these tools superimpose virtual clothing onto a consumer’s image or video in real-time. Gone are the days of relying solely on approximate measurements and static images before making a purchase; now, users can visualize how a product looks on them, shop with confidence, and indulge in a shopping experience that feels almost as authentic as trying clothes on in a physical store.

While promising, this use case carries notable risks. Virtual try-on applications may generate biased or inaccurate representations of certain demographics because of unrepresentative or limited training data. Retailers and brands need to thoroughly evaluate virtual try-on tools of interest and ensure that the training data used encompasses a diverse range of demographics.

Personalization and Consumer Experience Optimization

This subcategory is used as a catch-all for startups powering companies to pioneer personalized consumer journeys at scale. From AI stylists that customize product discovery through product recommendations, to video personalization platforms that can be used to ask for customer reviews, look to these startups to offer truly engaging shopping experiences online.

Customer Support and Sales 

Finally, this subcategory consists of startups that augment customer service capabilities by automatically resolving simple customer support queries and powering representatives to handle complex queries at scale. Examples of compelling features offered by these startups include chatting with customers in their preferred language, personalizing messaging using insights from customer profiles, and sharing relevant upsell recommendations based on a customer’s pain point.

For customer-facing text-based applications, retailers are encouraged to either constrain bot responses to queries or keep agents in the loop to verify AI-generated responses. Otherwise, open-ended bots can inadvertently make misleading or worrying statements to customers which could pose reputational risks.

In conclusion, generative AI technology holds immense potential to transform the retail industry. By automating tasks, personalizing customer experiences, accelerating innovation, and optimizing supply chain operations, retailers can unlock new levels of productivity and competitiveness. However, successful implementation requires strategic investments and a thoughtful approach to addressing ethical and regulatory considerations. The adoption of Gen AI has the power to reshape the retail landscape, thus enabling businesses to thrive in an increasingly digital and data-driven world.

Disclaimer:

This article was written with the support of GPT-3. While GPT-3 didn’t generate the text, it expedited the research and writing process — offering a taste of the future of work brought about by this new era of AI.

Are you building a company in the Gen AI space?

If so, please feel free to reach out to our team at Wittington Ventures. We’re actively looking to learn about new applications of the technology for industry verticals such as commerce, healthcare, and food.

Last but certainly not the least:

A special thank you to Ekam Sidhu for her contributions to this piece.