AI Shopping - The Holiday Tech Battle

This holiday shopping season, the battle for the future of e-commerce is being fought by Agentic AI assistants like Perplexity and Amazon's Rufus.

In this episode of the Inspiring Tech Leaders podcast, I discuss the debate of Innovation vs. Immaturity with AI Assisted Shopping and breakdown the following: 

💡 The Outdated Product Problem - Why generalist AI often recommends last-generation gifts.

💡 The strategic masterstroke of Perplexity's Instant Buy with PayPal.

💡 The power of Amazon's Rufus and its hyper-personalised, agentic utility.

💡 The legal conflict that is forcing the old guard to defend its turf and will define Retail Tech's next era.

The shift to Agentic AI is inevitable, but the technology is still working out the kinks, both technical and legal. Tune in to understand the true value proposition for Tech Leaders and consumers alike.

Available on: Apple Podcasts | Spotify | YouTube | All major podcast platforms

Start building your thought leadership portfolio today with INSPO.  Wherever you are in your professional journey, whether you're just starting out or well established, you have knowledge, experience, and perspectives worth sharing. Showcase your thinking, connect through ideas, and make your voice part of something bigger at INSPO - https://www.inspo.expert/

Everyday AI: Your daily guide to grown with Generative AI
Can't keep up with AI? We've got you. Everyday AI helps you keep up and get ahead.

Listen on: Apple Podcasts   Spotify

Chapters

• 0:00 Introduction
• 1:38 Innovation vs. Immaturity
• 3:04 The Contenders and Their Strategic Plays
• 5:01 The Pros, The Cons, and The Niche
• 5:02 [Ad] Everyday AI: Your daily guide to grown with Generative AI
• 5:50 (Cont.) The Contenders and Their Strategic Plays
• 7:25 The Battle for the Future of E-commerce
• 8:44 Wrap Up

Transcript

Welcome to the Inspiring Tech Leaders podcast, with me Dave Roberts.  This is the podcast that talks with tech leaders from across the industry, exploring their insights, sharing their experiences, and offering valuable advice to technology professionals.  The podcast also explores technology innovations and the evolving tech landscape, providing listeners with actionable guidance and inspiration.

In today’s podcast I’m discussing AI Shopping.  This time of year, the air is thick with two things, holiday cheer, and the frantic, often stressful, energy of holiday shopping. But this year, something has changed. The shopping experience is no longer just about you, the consumer, and the retailer. It is now about you, the retailer, and a new, highly sophisticated intermediary, the AI shopping assistant.

We are witnessing a full-scale invasion of the e-commerce landscape by Artificial Intelligence. In the last few weeks alone, we’ve seen major moves from OpenAI, Perplexity, and Amazon, all rushing to position their AI tools as your essential holiday shopping partner.

According to Adobe, AI-assisted online shopping is projected to grow by a staggering 520% this holiday season. That number is not just a statistic; it is a clear signal that the market is ready to embrace this technology. But the question for every tech leader, every consumer, and every entrepreneur listening to this: What is the true value proposition here? Is this a genuine leap forward in efficiency, or is it simply a new layer of friction designed to capture more of your attention and your wallet?

I believe the answer lies in understanding the difference between innovation and immaturity.

The initial promise of AI shopping is seductive: simply ask for what you want, and the AI handles the rest. But the early returns suggest a significant gap between the promise and the reality.

A recent tech article highlighted this perfectly, arguing that for all the hype, these AI assistants are currently stuck in the past.  The core problem is that AI’s consistent tendency is to recommend outdated products. 

This isn't a failure of intelligence; it's a failure of data and context. Generalist Large Language Models are trained on vast, but static, datasets. They struggle with the constantly evolving, real-time nature of product cycles. They default to products with the most reviews and the most historical data, which are, by definition, the older models. If you trust the AI implicitly, you could land up giving a last-generation gift.

This is the immaturity of the current generation. It’s a powerful tool, but one that still requires a human expert to validate its findings. It cannot yet replace the nuanced, up-to-date buying guide written by a person who understands the subtle differences between say an Apple Watch 10 and an Apple Watch 11. 

But this frustration is the very thing driving the next wave of development, this being Agentic AI. This is the shift from a conversational assistant to a digital proxy, where an AI doesn't just give you a list, but actively performs tasks on your behalf. This is where the true battle for the future of e-commerce is being fought.

Let’s look at the two most compelling strategic plays defining this space: the Challenger and the Incumbent.

First, the Challenger: Perplexity. Known for its powerful answer engine, Perplexity has launched its AI shopping assistant with a clear focus on user empowerment and complex query resolution. They allow you to ask highly nuanced questions, like "What’s the best camera for a motion designer focused on photo and video?" and generate a tailored response based on your preferences and past searches.

Crucially, Perplexity has partnered with PayPal to introduce Instant Buy. This is a strategic masterstroke. It allows the user to go from a complex question to a completed purchase without ever leaving the Perplexity interface. It removes the friction of being redirected to a retailer's site, a move that Perplexity frames as liberating the user from the ad-based model of the old internet.

Now, let’s turn to the Incumbent: Amazon, with its AI assistant, Rufus. Rufus is integrated directly into the Amazon ecosystem, and its power is simply unmatched in terms of data access. It is built on Amazon’s massive product catalogue, its community Q&As, and, most importantly, its account memory. Rufus leverages your individual shopping history to provide hyper-personalised suggestions.

Rufus is an agentic AI designed for utility. It can track prices, automatically purchase items at a target price, a feature that Amazon claims saves customers an average of 20% per purchase, and even process a photo of a handwritten grocery list and add the items to your cart. This is a tool designed to maximise convenience and conversion within the Amazon ecosystem.

The strategic takeaway is; the battle is for the seamless transaction. Whether it’s OpenAI with Shopify or Perplexity with PayPal, the goal is to close the loop. The AI that can most efficiently take you from intent to purchase, while providing a perceived value, will win the holiday season and beyond.

As tech leaders, we must weigh the true impact of this technology. For the consumer, the Pros are undeniable.

The first is utility. For known purchases, reorders, or price tracking, these tools are a massive time and money saver. Rufus’s ability to auto-buy at a target price is a genuine financial benefit. The second is complex query resolution. The AI can synthesise information faster than any human could, providing a starting point for research.

However, the Cons are equally significant, and they fall into two critical areas: accuracy and agency.

The accuracy issue, as we discussed, is the outdated product problem. For cutting-edge tech, the AI is still a risk.

The agency issue is more profound. Perplexity’s model, which learns your habits and preferences, comes at the cost of privacy. But beyond privacy, there is the question of human agency. If an AI is too effective, it can create a walled garden where you only see the products the AI deems appropriate, potentially eliminating the joy of discovery and the serendipity of shopping.

This is why the specialised startups are not worried about the entry of the giants. They argue that generalist AI lacks the domain-specific data needed for nuanced categories like fashion, interior design or specialised equipment. For example, finding that guitar with a great tone that matches your style of music is not the same as finding a television; it requires a level of emotional and contextual understanding that general LLMs simply don't possess. The future of e-commerce may not be a single winner, but a specialised ecosystem of vertical AI agents.

The philosophical debate over agency and friction has now spilled into the legal arena, giving us a clear view of the stakes.

Amazon recently issued a cease-and-desist letter to Perplexity, demanding that its agentic browser, Comet, stop making purchases on behalf of users in the Amazon Store.

Amazon’s argument is that Perplexity’s agent degrades the Amazon shopping experience. They claim the AI bypasses crucial engagement points, might not select the best delivery options, and prevents customers from seeing important product information. In short, it disrupts the carefully engineered flow of the Amazon shopping experience.

Perplexity’s response was sharp, framing the move as “Bullying is not innovation”, and positioning Amazon as a tech Goliath trying to protect its ad revenue and the friction of its traditional model. Perplexity argues that its agentic AI is about liberating the user from spam ads and predatory offers.

This is the core conflict: Is the friction in e-commerce a bug or a feature? Amazon argues it’s a feature, necessary for a complete, informed transaction. Perplexity argues it’s a bug, a barrier to be eliminated by agentic AI. This legal fight is a clear signal that the rise of agentic shopping is forcing the old guard to defend its turf, and the outcome will define how we all shop for years to come.

So, as you head into the final weeks of holiday shopping, what is my final thought

I’d use AI for utility, but not for discovery. Using Amazon’s Rufus to track prices and auto-buy selected items. Perhaps use Perplexity to quickly narrow down a complex search. These tools are excellent for saving time and money on known purchases.

But for the gifts that truly matter, the cutting-edge tech, or the highly personal items, the AI is still maturing. The risk of buying an outdated product or an irrelevant product is real. For those purchases, you still need to rely on human expertise and your own critical judgment.

The shift to agentic AI is inevitable, but the technology is still working out the kinks, both technical and legal. This holiday season is a massive real-world stress test, and I will be watching closely to see which models ultimately prevail.

Well, that is all for today. Thanks for tuning in to the Inspiring Tech Leaders podcast. If you enjoyed this episode, don’t forget to subscribe, leave a review, and share it with your network.  You can find more insights, show notes, and resources at www.inspiringtechleaders.com

Head over to the social media channels, you can find Inspiring Tech Leaders on X, Instagram, INSPO and TikTok.  Let me know your thoughts on AI Shopping Assistants.

Thanks for listening, and until next time, stay curious, stay connected, and keep pushing the boundaries of what is possible in tech.