AI trust gaps and misleading output
The hosts opened with data from a Deloitte report cited by eMarketer: 33% of US Gen AI users have experienced inaccurate or misleading output. Brian noted that when you interact with a chatbot like an expert, you tend to believe it more than a traditional search result. Andy compared it to talking to your grandfather: we assume good intent even when accuracy is uncertain.
The conversation turned to what happens when answer engines start accepting advertising. At that point, users will not know if the recommendation is based on aggregation or sponsorship, and the trust gap will widen.
Amazon Business context
Brian provided context on Amazon Business: Bank of America estimates 83 billion dollars for 2025, nearly five times Grainger and approaching AWS scale. Millennials and Gen Z now represent 75% of the workforce, and this Amazon-native cohort increasingly uses the platform for business purchasing.
Sixty percent of millennials say Amazon is their most preferred shopping channel. Two thirds make half or more of their online purchases from Amazon.
Brian Beck, Master B2B
Andy noted that Amazon recently signaled openness to allowing AI crawlers to index its site, reversing an earlier position that had some worried about a walled garden approach.
The debate setup
The main topic was a live debate at B2B Online in Atlanta. Brian led team 3P with Adrienne Hartman from JJ Keller and Samantha Schwarz from Midland Industries. Andy led team 1P with Shalin Shah from UPS and Rudy Abitbol from AMG Medical. Three rounds covered channel control, operational preferences, and financial returns.
Round one: channel and brand control
The 1P team argued that selling to Amazon gives you the buy box and keeps unauthorized sellers at bay. The 3P team countered that you control your own pricing and get more data. The audience vote was a tie at 95-95.
Round two: operational preferences
The 1P team argued that shipping pallets and trucks to Amazon is more consistent with how manufacturers operate. You do not handle returns. The 3P team noted that Fulfilled by Amazon handles the last mile, but acknowledged there is more to do on the seller side. The 1P team won 89-88.
Round three: financial returns
Brian opened by noting that 1P yields wholesale revenue while 3P yields retail revenue. On a per-unit basis, 3P is almost always better unless your product is very low priced or bulky. When Brian asked the audience who voted for 1P on financial returns, there was silence.
There was absolute silence. I said let’s do that again. And it was zero.
Brian Beck, Master B2B
The final score was 100-0 for 3P, the most decisive margin in debate history. The overall winner after a clap-off was the 3P team, 98-96.
What practitioners think
In a LinkedIn poll, 66% said 3P is the better approach for B2B companies on Amazon. Brian noted that the right answer is company dependent. Some clients do great on 1P. But the financial math generally favors 3P for products with reasonable average order value and shipping economics.