There’s been a lot of talk recently about agentic AI and its potential to transform industries. AI, combined with autonomy, is likely to reshape the way we live, work, and do business.
Larger organizations may have the resources to deploy autonomous employees and AI-driven cybersecurity, but smaller businesses don’t have to be left behind. The rise of cloud platforms offering agentic services means there are opportunities for all.
E-commerce, often highlighted as a sector primed for agentic AI adoption, is an industry built on digital interactions. Agents capable of handling shopping-related tasks, optimizing supply chains, and creating personalized customer experiences are already here.
Retail, in particular e-commerce, has been the poster child for agentic AI and is a sector where there is a lot of hype but also some very compelling use cases. So, let’s explore what’s happening in this space and what we can expect to see in the future.
What Is Agentic AI And Why E-Commerce?
First, just in case you’re not totally sure what I’m talking about – agentic AI refers to AI systems built around the same language processing capabilities as services like ChatGPT but capable of carrying out much more complex, goal-oriented tasks with minimal human intervention. For example, they can interact with websites, fill out forms, collect and submit information and make payments.
There are a number of reasons why e-commerce as a business sector makes a great testbed for agentic adoption.
Firstly, its infrastructure, business processes and operations are built around digital platforms and data. This means AI agents can more easily connect with, talk to and operate its systems.
Secondly, it’s a sector where battles are won and lost based on customer experience. Agentic AI could be hugely useful for e-commerce businesses looking to differentiate themselves based on more efficient, personalized and automated CX.
Today, AI agents can automate e-commerce inventory management, shipping and delivery, customer service and dynamic pricing strategies. Tomorrow – for businesses that clear the substantial technical, regulatory and ethical hurdles – they could be capable of far more.
Agentic AI In Practice: E-Commerce
As you would expect from the ripe possibilities, there are already many use cases for agentic AI in e-commerce.
Shopping bots are becoming available that are capable of hopping between online retailers and searching for the best prices, deals and promotions for their users. Salesforce has released its Agentforce platform, which lets retailers build AI agents that can interact with their systems.
In customer service, agents can help customers troubleshoot problems far more flexibly than standard genAI chatbots. They can even take control of users’ systems to diagnose and fix problems automatically. Amazon’s Q platform lets businesses get started with developing this agentic capability.
Agents can also help retailers dynamically price products and services by tracking the competition. They could even come up with novel new metrics that they could use to set prices more efficiently, taking into account factors humans might overlook. Amazon already gives its merchants the ability to have AI dynamically set prices, so adding agentic functions for even more powerful automation is a no-brainer.
And with AI and generative AI already deployed in supply chain and inventory management by companies like Walmart and Alibaba, agentic supply chain and inventory management bots are the next step. They will negotiate with suppliers and distributors to coordinate deliveries, restocking and optimizing the use of warehouse space.
Why The Future Of E-Commerce Is Agentic
Looking further ahead, I believe agentic AI will be just as integral to e-commerce as previous transformational waves, like machine learning and the mobile internet.
It could even turn out to be a significant step towards the long-term goal of developing generalized AI – AI that can learn how to do just about any task, much like a human, rather than just the tasks it was created for.
(For fictional examples, think of Star Trek’s Data or, more scarily, The Terminator, for examples of how awesome or terrifying that could be!)
While agentic and generalized AI are not the same thing, today’s agentic tools give us a glimpse of how that future could play out.
Businesses that want to be a part of this future will have to overcome technological, cultural, ethical and regulatory challenges. Perhaps the biggest challenge will be working out how this new breed of machine workers fits in with human talent.
But with agentic AI set to define the next stage in the evolution of e-commerce, any investments in conquering those hurdles are likely to be more than worthwhile.
