Fiftytwo's Blog - Innovative IT-solutions for retail

What AI means to retail: exploring opportunities and challenges

Written by Rasmus Bach Hansen | 29. October 2024

Step into a modern store. AI has been working behind the scenes, from personalized shopping inspiration before you even visit, to selecting your first item and checking out. AI drives faster payments, customizes product suggestions, and optimizes stock levels. Retailers are using AI to create smoother, more efficient, and tailored shopping experiences.

As discussed by Thomas Grane, expert in AI and former CIO of Matas, during Retail Dialogues 5: "In this day and age, it is really just pattern recognition and rules on steroids. AI’s role is to support human decision-making, not replace it."

AI’s role in retail: supporting better decision-making, not replacing people

Retailers are continually challenged with tasks like managing inventory efficiently and meeting growing customer expectations. AI has become essential for optimizing these processes, but its true strength lies in supporting human decision-making, rather than replacing it.

"AI in retail helps speed up processes and improve accuracy, but the human aspect is vital. The strategic decisions remain in human hands."
– Thomas Grane  

AI systems can quickly analyze large amounts of data, offering insights that would be difficult to obtain manually. This enables retailers to focus on making strategic decisions, while AI handles tasks such as detecting fraud at checkouts or analyzing customer preferences to improve marketing efforts. 

Real-world applications of AI in retail

AI is already making a noticeable impact on how retailers operate, from speeding up checkouts to offering personalized product suggestions. These changes aren’t abstract concepts; they’re happening now in stores, restaurants, and large venues. Let’s dive into some real-world examples.

Fraud detection at self-checkouts

Self-checkout fraud is a significant challenge for retailers. In collaboration with Nomitri, Fiftytwo integrates the EmSCO solution, which uses on-device AI to detect fraud in real-time. This system identifies missed or fake scans, provides article recognition, and even performs age estimation, all without needing cloud or Wi-Fi. By running on edge devices, EmSCO ensures faster, more reliable performance.

The system promptly alerts staff to potential fraud, allowing for immediate intervention. This approach reduces losses and enhances the customer experience by providing feedback when scanning errors occur, such as obstructed barcodes. This collaboration between Nomitri and Fiftytwo highlights how AI can improve both security and the checkout experience while requiring minimal infrastructure.

Frictionless checkout systems in retail, QSRs, and stadiums

Frictionless checkout systems powered by AI reduce wait times by eliminating the need for traditional checkout processes. In these systems, customers scan their payment method upon entry, select items, and leave without manually scanning products or standing in line.

At Parken Stadium in Copenhagen, fans could purchase items without needing traditional checkout lines during events like UEFA Champions League matches. Similar systems are in place at venues like Mercedes-Benz Stadium in Atlanta and Ford Field Stadium in Detroit, where frictionless checkout reduced transaction times by 50% and boosted sales by 170% ​(PYMNTS.com)​(AiFi).

This technology is proving valuable in fast-paced environments such as quick-service restaurants, music festivals, and large events where crowds gather. Whether at a stadium, fast-food outlet, or big event, these systems ensure customers can make their purchases quickly without missing important moments, be it a game, a concert, or grabbing a meal on the go.

AI in unattended stores: convenience meets safety

Unattended stores are becoming more popular, and AI plays a crucial role in balancing convenience with safety. AI systems monitor customer behavior, and in emergencies, such as a customer collapsing, these systems can immediately alert remote staff, ensuring timely intervention.

In Denmark, Salling Group opened føtex GO!, the country’s first fully cashier-less store, where customers shop 24/7 without staff. Through our partnership with StrongPoint, 52ViKING POS plays a role in the solution. Advanced camera technology tracks what customers pick up, and payments are processed automatically as they exit.

AI is not only optimizing retail operations but also enhancing the efficiency and security of unattended stores. By monitoring activity in environments without physical staff, AI ensures that both convenience and safety are prioritized. These systems help manage stock levels, notify suppliers when inventory runs low, and provide accurate transactions while maintaining high service standards.

Optimizing inventory with AI: efficient stock management

Managing inventory efficiently remains one of retail’s main challenges, but AI-driven solutions are helping by providing real-time insights and predictive capabilities. AI systems continuously track stock levels and can automatically reorder items when stocks run low.

"Camera technology can be used for stock monitoring. It can recognize when shelves are out of stock and can even trigger automatic reordering of products, extending the supply chain and improving demand forecasts."
– Søren Brammer Riis

An example is Vusion’s AI-powered computer vision technology, Captana, which uses machine learning (ML) algorithms to analyze shelf conditions in real-time. The system processes camera visual data to detect out-of-stock items and identify discrepancies with stock records. When anomalies, like empty shelves or misplaced products, are detected, the AI system instantly alerts staff for immediate replenishment. This not only ensures product availability but also minimizes waste and prevents overstocking. As a result, retailers can adjust their supply chains based on real-time demand patterns, improving efficiency and customer satisfaction.

Personalized product recommendations at Matas

Thomas Grane played a crucial role in developing an AI-driven personalization system that analyzed over 50 billion relationships between customers and products, delivering highly targeted product recommendations. This solution led to a 300% increase in conversions for Matas.

"At Matas, predictive AI helped increase conversion by 300%. We used AI to analyze customer behavior, finding the most relevant products and injecting them into personalized email marketing. This resulted in a boost in sales."
– Thomas Grane

By leveraging AI insights, the system provided customers with relevant product suggestions, strengthening customer loyalty and increasing sales. This is a prime example of how harnessing data and AI can deliver powerful results when opportunities are identified and acted on.

Ethical AI in retail: balancing transparency with customer trust

“When you implement AI, you need to consider how it affects customer experience and your company’s DNA."
Thomas Grane

As AI becomes more widespread, ethical considerations are becoming critical. AI systems can be prone to biases if not adequately monitored. Thomas Grane highlighted the need for "keeping humans in the loop" to ensure that AI models remain fair and unbiased.

"While AI can automate processes, human oversight is essential to ensure it aligns with company values and treats customers fairly," he added.

Retailers must also comply with regulations like the EU AI Act, ensuring that AI systems are transparent and protect consumer privacy.

Best practices for ethical AI in retail

  • Human oversight: People should review AI decisions, significantly when they impact customer interactions or pricing.
  • Bias testing: Regularly test AI systems for potential biases to ensure fair treatment of all customers.
  • Transparency: Clearly communicate how AI is being used to enhance customer experience, building trust through openness.

AI is the tool, but retailers drive the change

"AI improves speed and accuracy in retail, but it’s the human strategy that drives success."
– Thomas Grane

AI’s benefits extend beyond tasks like fraud detection, inventory management, and personalization. Its true value lies in helping retailers make better decisions, streamline operations, and enhance customer satisfaction.

However, its effectiveness depends on how thoughtfully it’s used, with human oversight remaining essential. When applied strategically, AI can reduce costs and support smarter business choices without losing the personal touch that matters to customers.

Retailers who use AI wisely will be best positioned to succeed without losing sight of their strategic goals. How will your business make the most of it?