A week in AI is like a year in other industries. I hope these issues become your weekly source of AI information, inspiration, and ideas. If we haven’t met before, I’m Amanda Smith. I write about AI and the fascinating folks who are building in this brave new world.
Good morning. I’ve tried using ChatGPT Search before with lackluster results. I was trying to find an organic version of an old lipstick color I loved.
If only I knew about this week’s AI product. It’s built for complex queries, just like this. This female founder was building an AI agent well before that term even existed.
Let’s dive in.
🤝 This edition is brought to you by Lightfield
“Tell me why we keep losing to [competitor] and what to change.”
“Find every churned account that mentioned the feature we just shipped.”
“Build a pipeline report for my board deck.”
These are questions founders ask Lightfield every day.
It captures every email and call recording, then lets you run real analysis on the data — not dashboards, not filters, but real natural-language queries that return answers with citations from actual conversations.
Set up in 3 minutes - just connect your email and calendar and watch your CRM build itself.
Try Lightfield → use code AIODYSSEY13 for 3 months free.
This week in AI:
Anthropic files lawsuit to block Pentagon blacklist
The battle with the U.S. military over usage restrictions continues. The AI giant is fighting the Pentagon from placing it on a national security blacklist.
Google’s “Ask Photos” search feature complaints
User complaints caused Google to return to “classic” search experience over newer “Ask Photos.”
Anthropic releases Labor Market Report
An important read, if you haven’t seen this yet.
Company background: Wizard
Founded: September 2021
Team size: 39
Funding to date: $50 million
ARR: Not disclosed
Growth metric: Nailing their positioning as a specialist agent, but without being constrained by a catalogue.
Melissa Bridgeford, Wizard
Back in 2017, Melissa Bridgeford seeded a startup in what was then called the conversational commerce space. In 2021, she teamed up with the former president and CEO of Walmart e-commerce, Marc Lore, who was also bullish on AI shopping agents, to build Wizard.
Their shared vision to fix the broken e-commerce experience that overwhelmed consumers – “an explosion of products and an implosion of trust.”
This was pre-ChatGPT, before “AI agents” had captured culture.
The agentic commerce era
Consumers need to ingest and analyze so much data today in their decision-making process– ratings, reviews, editorial, social signals, and Reddit rabbit holes. Bridgeford’s vision for Wizard was to help consumers shop in seconds, with a trusted AI agent that knows you better than yourself.
She predicts agentic commerce is going to be bigger than the shift from brick-and-mortar to online. Consumers will get their AI agent to search the entire commerce layer for the best products, pricing, and purchase decisions.
The first pillar was the AI agent delivering smarter search. The second pillar is executing seamless checkout within Wizard, where an order gets routed to the participating retailer. The third pillar she sees coming is personalization, where the agent becomes proactive and predictive.
“For example, it might say, hey we know that your husband’s birthday is next month. Here’s three things that he’s going to love,” she said.
The fourth pillar is where the agent transitions from assisting to executing purchases for you.
Building in beta mode
Wizard spent five years in beta mode, long before AI agents were commonplace terms. They launched publicly last month. Bridgeford said the bar is tremendously high, when dealing with consumer and merchant money. “You must get the product and inventory information right, as well as the pricing.”
They knew it was the right time to launch when they had high demand from consumers, merchants, and payment companies within the agentic commerce race.
Wizard spans across anything you can buy and everything on the web, except groceries. They’ve focused on “playbook categories” which represent 80% of e-commerce volume and consumer queries. This includes electronics, makeup and beauty, kitchen and dining, apparel, footwear, home and toys.
Bridgeford said you can click on the PDPs for a rich AI analysis of each product Wizard recommends. “People want to know the commentary on a specific product, whether it’s on TikTok, Instagram or Reddit,” which is on their future roadmap.
Partnerships & business model
Their partnerships span across both merchants and payment companies, such as a data and checkout integration with Best Buy. “The Best Buy CEO announced their partnership with Wizard, alongside OpenAI and Google in their strategy for agentic commerce,” Bridgeford shared.
Wizard is rolling out additional checkout integrations across major retailers and categories to create that universal cart experience.
“On the payment side, we’ve partnered with Stripe to join their agentic commerce protocols, which will help merchants on board at a much faster rate. Instead of customized one-off checkout integrations, you're able to really scale that checkout integration much faster for merchants, which helps us deliver the broader merchant checkout portfolio to our consumer base.”
Wizard’s business model is focused on a take-rate and affiliate rate from participating merchants.
Bridgeford explained how there are specialist agents on a single marketplace (like Amazon’s Rufus or Walmart’s Sparky) that are constrained by their catalogue and full of sponsored ads. And then there are generalist agents, such as ChatGPT or Google, that are focused on horizontal capabilities.
She believes specialist agents that solve the complexities of e-commerce will win.
Key marketing channels
Wizard’s growth levers include:
Merchant partner activations
Referral loops with active users (Wizard advocates)
Influencers (trust-based recommendations to help educate users around AI shopping agents and how to interact with them)
The education element in their content is key, as search is evolving two one or two keywords to complex queries. “Agents are training consumers to ask longer questions and engage more conversationally. In Wizard, our queries are over six words, three times the length of traditional keyword searches. Instead of just saying SPF, you might say SPF that is organic, under $30, and good for kids. We’re able to deliver the best product recommendations with a higher, more intense search.”
Takeaways
Rapid product iteration is table stakes today. Wizard focused on rapid product learning during beta mode to optimize performance and UI.
Pioneering a change in consumer behavior requires a focus on education marketing. Competitors can also support consumer education.
Getting it right is better than being first. Wizard waited until they had a product with very high accuracy before they brought it to the market.
How's the depth of today's edition?
If one of these stories stuck with you, I’d love to hear which one.
Speak soon,
Amanda
P.S. If you want to get a founder feature about your own story, reply to this email. If you’d like to reach our newsletter audience (founders, creators, and marketers), click the button below.
If you’re new here, I’m over the moon you’ve joined us! To help me craft content that’s actually useful (and not just noise in your inbox), I’d love it if you took 1 minute to answer this quick survey below. Your insights help shape everything I write.
✨ Insane Media is more than one voice
Dive into our other newsletters - where psychology meets the creator economy, e-commerce marketing, and Human resources.
