AI News

OpenAI 在 ChatGPT 内推出 Businesses 测试版,品牌可建档案并配置购物智能体

The Rundown AI更新于 11月11日 22:30

OpenAI invites brands to set up shop in ChatGPT | The Rundown AI

AI 导读

OpenAI 在 ChatGPT 中推出 Businesses 测试版,品牌可认领认证档案、查看 ChatGPT 提及自己的频率,并搭建面向买家的 AI 智能体。档案目前对符合条件的公司免费开放,智能体先面向美国线上零售商和数字服务公司上线;品牌可自行撰写智能体指令并补充产品信息,买家从档案进入独立标注的对话。后台面板还展示提及次数、跳转点击和发现话题,品牌也可付费让智能体成为 ChatGPT 广告的落地页。

推荐理由

AI 评分 68

OpenAI 把品牌档案、提及统计和购物智能体放进 ChatGPT,读者可据此判断品牌在 AI 搜索里的曝光方式如何变化。

主题

标签

正文

AIOpenAI invites brands to set up shop in ChatGPT

PLUS: Put your open Chrome tabs to work with AI

Zach Mink

Good morning, AI enthusiasts, and welcome to our 5,653 new readers. Google conditioned companies to obsess over their search listings for decades. That same scramble has been playing out on AI platforms, and OpenAI just gave brands an official way in.

Its new ‘Businesses’ beta gives brands a verified profile, a view of their mentions, and a personalized AI agent for direct buyer questions, opening a new way to meet customers where they’re increasingly searching… inside ChatGPT.

In today’s AI rundown:

OpenAI invites brands to set up shop in ChatGPT

Anthropic stretches Claude across the office

Put your open Chrome tabs to work with AI

‘AI 2027’ authors move up their timelines

LATEST DEVELOPMENTSOPENAI

🛒 OpenAI invites brands to set up shop in ChatGPT

Image source: OpenAI

The Rundown: OpenAI just introduced Businesses in ChatGPT, a beta hub where brands can claim a verified profile, track how often the chatbot mentions them, and set up AI agents for shoppers, which paid ads can now point to.

The details:

Profiles are currently free to claim for eligible companies, while agents are live first for U.S. online retailers and digital service firms.

Brands can write their agent’s instructions and add product details, with shoppers opening it from the profile in a labeled chat separate from ChatGPT’s answers.

A dashboard shows how often ChatGPT mentions a brand and drives clicks to its site, plus “discovery topics” that can point to gaps in what the brand shares.

OpenAI first piloted ad-linked agents in September, and brands can now pay to make their agent the landing spot for ads in ChatGPT, with each paid spot labeled.

Why it matters: Brands used to fight for a spot in Google’s results, and in the AI era, that’s shifted to getting name-dropped by the assistants themselves. Companies were mostly guessing what ChatGPT said about them or paying outside specialists to find out. OAI just handed them that visibility directly, plus a one-on-one line to potential customers right inside the chat. TOGETHER WITH MEMOKET

🗓️ Your conversations, now in Apple CalendarVisit memoket.ai

The Rundown: Memoket Gem is an AI wristband that captures conversations with one press and turns them into summaries, action items, and connected memory. Now with Apple Calendar integration: connect in the Memoket iPhone app to automatically sync new recordings as calendar events.

With Memoket, you can:

Find recordings by when the conversation happened

Revisit decisions and context across conversations

Bring relevant memories into ChatGPT and Claude

Use the app on your Apple Watch, coming this month

Discover Memoket Gem. ANTHROPIC

🗄️ Anthropic stretches Claude across the office

Image source: Anthropic

The Rundown: Anthropic just rolled out Claude Sheets and Whiteboards in beta — spreadsheets and shared boards Claude builds alongside a team — while making its Claude Data analytics tool generally available and announcing Claude Meetings, an assistant that sits in on video calls.

The details:

Sheets can build a working spreadsheet with formulas via prompt and export it to Excel, while Whiteboards maps a plan onto a shared board for a team to edit.

Claude Data, originally named Dashboards when it entered beta in October, turns company metrics into narrated videos and visual data stories.

Product lead Nicholas Lin said Claude joins nearly every Anthropic meeting after months of testing, with a Google Meet beta coming to Team and Enterprise users.

Upgraded connectors, the links between Claude and other work apps, now also open tools like Adobe and Atlassian fully inside Claude without switching platforms.

Why it matters: Claude continues to eat both legacy office software and the newer apps that followed, with spreadsheets, dashboards, and meeting tools now joining Slides, Design, and Docs. The apps it hasn’t rebuilt now open inside Claude anyway, pulling more use into the AI assistant and away from the tools the workday used to run on. AI TRAINING

🌐 Put your open Chrome tabs to work with AIVisit The Rundown University

The Rundown: In this guide, you’ll use ChatGPT’s updated Chrome extension to compare open pages, check the facts, and group the tabs you need. Start in Chat to read, then use Work for a browser action you approve.

Step-by-step:

Install the official ChatGPT extension from OpenAI and open its side panel. Use a practice Chrome profile with public pages for your first run

Open three pricing pages and tell ChatGPT which tabs to read. Give it your team size, required features, budget, and billing period

Check the comparison against the pages. Confirm the full team cost and leave missing details unverified rather than guessing

Switch to Work where available and ask it to group only those three tabs. Review the list first, then check the actual group in Chrome

Ask for a short recommendation with source links and one question to resolve. Keep trials, purchases, and messages outside the task

🎰 Only 1 in 5 AI use cases pays off Visit gartner.com

The Rundown: Enterprise AI use cases grew by 400% this year, but not all are delivering ROI. Investment without value only creates bad momentum. These Gartner IT Symposium/Xpo takeaways show what separates mere AI activity from real, measurable results.

What you’ll learn:

How to build momentum that compounds value

Ways to prevent key risks

How to make people your AI safety net

Download the insights. AI RESEARCH

📆 ‘AI 2027’ authors move up their timelines

Image source: AI Futures Project

The Rundown: The AI Futures Project, the forecasting group behind AI 2027, just moved up its forecast for AI that matches the best human AI researchers up to 2027 — citing AI evaluator METR’s recent outside review of Anthropic’s AI speedup as its biggest reason.

The details:

The group’s paper published in 2025 depicted AI rapidly improving itself, but most of the authors later pushed their dates for AI that beats top experts into the 2030s.

METR estimated Anthropic now improves about 2x as fast, mostly thanks to AI, a pace the forecasters’ models gave under 4% odds of seeing this early.

Ex-OpenAI researcher Daniel Kokotajlo gave the earliest date, a 50-50 chance by May 2027, while the other five forecasters land between August and November.

Kokotajlo wrote on X that he is “scared, sad, and angry,” and the group says that if its forecast holds, a pause “should be the top priority for the AI industry.”

Why it matters: AI 2027 is considered one of the more prescient forecasts for how the industry has accelerated, and even its own authors said the pace shown in METR’s report caught them off guard. If their updated projections are as accurate as their old ones, the industry has months, not years, to prepare for the implications of superhuman AI research capabilities.

▸ Polly built a caregiver recordkeeping app with Claude

Today’s workflow comes from reader Polly Bouker:

“My son is disabled and receives state funding for caregiving. The program requires daily records of what was done, whether he met his goals, and any behavioral or health issues. A program coordinator reviews these periodically, and an auditor may review them too. I used Claude to write the code for a web app built with Google tools that handles this recordkeeping.

Each day, caregivers open the app and check off each item on my son’s daily task list, which he completes with their help. The app writes the required daily summary from those check-offs and adds goal tracking. Everything goes into a printable report for the coordinator or an auditor.

My son has his own view showing how much of his list is checked off and how close he is to his favorite reward. Because he is motivated to see that he has earned the reward, he reminds the caregivers to complete their part.”

See the workflow on our Community Workflow Hub →

🛠️ Trending AI Tools

🛍️ Businesses in ChatGPT - OpenAI’s new brand profiles with product listings and chat agents

🎆 MiniMax H3-Image - New open-weight model that creates and edits 4K images

🎨 Adobe Firefly - AI studio adding beta tools for design, video, and workflows

📰 Everything else in AI today

Triple Whale’s 50 LLM Prompts to Crush BFCM 2026 turns your data into answers with copy-paste prompts for any LLM: 35 to plan, 10 to run live all weekend, 5 to save soft results.*

Anthropic opened the investor pitch for its IPO coming next week, also saying in an updated filing that it now considers its AI R&D capability threshold met, based on what it expects from upcoming models.

Adobe previewed MCP connector tools that let users’ own AI agents edit Photoshop, Illustrator, and Premiere files coming in early 2027, plus Firefly Graph, a beta builder for repeatable, shareable AI workflows.

France unveiled the Paris Call for AI Safety and Security, with signers such as OAI, Anthropic, and Mistral committing to build emergency off switches into their most advanced models.

OpenAI acquired Odessia, a months-old AI trip-planning startup from Sonder founder Francis Davidson, who is joining OpenAI to work on travel.

*Sponsored Listing

🎓 Highlights: News, Guides & Events

RSVP to our next workshop on Nov. 12: Use Claude to its full potential

Anthropic says ‘pacing, not pausing’ with Fable 5.2

Jensen Huang admits affair amid spy claims

Zach Mink • 8 minutesAI

Rowan Cheung • 4 minutes

Stay Ahead on AI.

来源

The Rundown AI