Agent Beat for September 20, 2026

Welcome to Agent Beat, a daily briefing on how AI finds brands and how agents choose what to buy. I'm your host. Today is Sunday, September 20, 2026.

A CyberAgent survey says AI search usage in Japan passed 50% for the first time. ChatGPT is the most used tool, and Gemini is the fastest growing. The survey also found 47% of respondents said they moved more than half of their searches to AI. Marketers should treat AI assistants as a primary discovery surface and track brand mentions across ChatGPT and Gemini, not just traditional search results.

That shift is changing where advertising can appear. According to the Mobinside column, TIME serves a second version of its site that only AI crawlers see, and it sells ad space on that version. Ally Bank and the Project Management Institute are the first advertisers. TIME says bots now outnumber human visitors on most days. Marketers should expect a market for placements that only models read, and should write claims they are happy to have quoted back when someone asks a chatbot which brand to pick.

Tools are adapting to that reality. Ahrefs increased the monthly API units available for AI tools and expanded analysis of conversations with ChatGPT and Claude, according to the Japanese report. The change gives marketers more room to run competitor and keyword research inside AI workflows. Marketers using AI assistants for SEO should expect to pull more Ahrefs data per month and to analyze AI conversations for AEO.

Agencies are building services for the same shift. Dentsu Digital started a service to help brands prepare product data so AI agents can understand and recommend products, and to test buying paths. The company says the service is aimed at an era when AI agents buy on a user's behalf. Marketers should review whether their product feeds and structured data are ready for agent-mediated shopping, not just human storefronts.

That idea extends to the software layer. Stratechery argues Salesforce is giving up user interface as a competitive moat because interfaces are disappearing. The column links that shift to agents becoming the interface and to headless commerce. Marketers selling through agent interfaces should expect product data and checkout to matter more than a branded storefront.

If you are deciding where to track AI visibility, Semrush reviewed nine AI search engines for 2026. The post is meant to help marketers choose where to track or earn AI visibility. Marketers should use the comparison to decide which answer engines to monitor first.

Google is also testing changes that could affect traffic. According to SEO Südwest, Google is preparing a test of a new button in Discover that opens AI search. The button could send fewer clicks to the sites named in Discover. Marketers should watch whether Discover referrals drop if the test expands.

And if you publish research, SEO Südwest notes that arXiv.org is one of the sources ChatGPT prefers. The outlet asks whether publishing research there can improve AI visibility. Marketers with technical or research content should test whether putting work on arXiv changes how often ChatGPT cites them.

Tools for optimizing citations are emerging. LinkShare Japan added a feature to its LinkSurge tool that diagnoses how suitable each heading block is for citation in AI search and suggests improvements. The tool is meant to help optimize articles for AI citations. Marketers should consider checking headings and content blocks for AI citation potential, not just for classic ranking keywords.

Platforms are reporting early adoption. HubSpot launched AEO in April, both inside Marketing Hub and as a stand-alone product. CEO Yamini Rangan said, "We launched HubSpot AEO in April, both within Marketing Hub and as a stand-alone product. And since then, 32% of Marketing Hub Pro+ customers have activated AEO and nearly 16,000 customers have activated a stand-alone AEO trial in Q2." Rangan also said the agents now in market include "AEO agents to show up in LLMs," putting AI-search visibility among HubSpot's agent products and trials.

Adobe is seeing similar traction. Adobe launched its first unified brand visibility solution, combining Adobe LLM Optimizer with Semrush's AI visibility intelligence for generative engine optimization. Management said paid customers for brand visibility solutions doubled quarter over quarter. President of Digital Experience Anil S. Chakravarthy said, "We are seeing tremendous customer interest in tapping our database of nearly 300 million real world AI search prompts to understand how their brands are showing up across leading platforms such as ChatGPT, Google AI Mode, Microsoft Copilot, and Perplexity AI." Adobe is selling visibility across generative search engines on top of its existing web content management, with the offering already released and drawing customer interest.

Yext is also moving into the space. CEO Michael Walrath said, "In June, we acquired GoShine, a brand visibility platform built for the AI search era." Walrath also said, "The engines now compose a single answer in place of that list, and they build it from whatever data and content they trust, which is why a brand's pages can rank well while the answer recommends a competitor." For Yext, the call ties its AI-search and agent-traffic efforts to GoShine/Brand Scout and the Corvo AI prototype, with GoShine piloted with a small number of enterprise customers before broad availability later this year and acquired revenue not expected to be material this fiscal year.

That's your briefing for Sunday, September 20, 2026. Thanks for listening to Agent Beat. For the full written brief, go to agent-beat.com.