ChatGPT Ads Update: an Ads Manager Plugin, Looser Custom Audiences, and a Euro Minimum
OpenAI's 4 September product email adds a plugin that builds campaigns from ChatGPT or Codex, loosens custom audiences, widens conversion matching, reports carousel cards and paces total budgets. The same week its Help Center quietly published a 15 EUR daily minimum.
Guilherme Hortinha · 7 September 2026
Can your business run ChatGPT Ads? Check in 10 seconds.
OpenAI's weekly product email of 4 September 2026 shipped seven changes to ChatGPT Ads: an Ads Manager plugin for ChatGPT and Codex, looser custom audiences, more conversion matching fields, card-level reporting for the product feed carousel, pacing for campaign-total budgets, an announced view-through optimization model, and a restatement of the Europe, India, Middle East and North Africa rollout. The change that matters most to a European advertiser was in none of them: the Help Center's minimum daily budget table went from ten currencies to 22 the same week, and it now has a euro row. 15 EUR a day.
The plugin: campaigns from a sentence
The headline item is a ChatGPT Ads Manager plugin. Installed from Plugins in ChatGPT or Codex and connected to an advertiser account, it creates, manages and analyzes campaigns from plain language: turn a website or a brief into ready-to-launch ads, create variants, update campaigns, troubleshoot delivery, review performance, get prioritised recommendations. OpenAI's own qualifier is that the plugin "previews recommended changes and asks for confirmation before applying them", so nothing goes live from a sentence.
Two things the coverage skipped. There is no Help Center article for the plugin yet; the ChatGPT Ads collection lists 27 articles and none of them mention it, so the email is the only OpenAI wording. And OpenAI's plugin documentation says installing or using any individual plugin "depends on your plan, workspace, role, region, and its included capabilities". Whether your account can see it is a separate question from whether it exists.
Our read: the structure it produces is the same three-level structure a bulk upload needs, and the same validation rules decide whether its output launches. The parts that still need a person, which situation each ad group targets and whether a hundred ads are actually distinct, are the parts that decide cost. Where it fits is in the bulk upload guide.
Custom audiences, loosened
Custom audiences arrived in July 2026 with a hard rule: 25,000 matched users or nothing. The rule now has edges.
Members can be added, removed or replaced without recreating the audience, and the campaigns and ad groups using it keep working.
One upload can mix identifier types: email, phone, hashed email, hashed phone and, new, Google Advertising ID.
Audiences under 25,000 matched users can be used for exclusions. They still cannot be included or given a bid multiplier.
Audiences can exceed five million members.
Size ranges are more granular.
The Help Center article, updated on 5 September, also gives the numbers behind the rule: 100,000 matched users recommended for inclusion or bid adjustments, files up to 500 MB, processing in 20 to 30 minutes, uploaded files deleted within 24 hours, multipliers from 0.1x to 10x with the highest winning when a viewer is in several lists. We have written it up as a guide to custom audiences, including the part OpenAI does not say: for a services business with a few thousand contacts, exclusion is the only mode the floor allows, and it is the useful one.
More fields to match a conversion on
The Measurement Pixel now takes hashed phone number, first name, last name, region and postal code alongside the hashed email and external ID it already accepted. The Conversions API takes plural list fields for hashed email, phone, external ID, first and last name, country, city, region and postal code, plus the Android advertising ID for mobile. Every personal identifier travels as a SHA-256 hash of the normalised value, and the developer page carries the normalisation rules.
More fields means more conversions tied back to an ad, which is what every advertiser wants. It also means automatic advanced matching now works with more of your form data. The opt-out is unchanged and is covered in the conversion tracking guide.
The carousel is serving, and now it reports
On 5 August OpenAI said it had "started testing" a multi-product carousel for product feed campaigns. On 4 September it started reporting on it: card impressions and clicks in Ads Manager, product-level fields in the Insights API, complete data from 20 August 2026 on a rolling 30-day basis. A carousel card impression is counted when an individual product card becomes viewable, and it is separate from the billable ad impression.
You cannot report card-level clicks on a format that is not live. The carousel has left testing, for feed campaigns, and OpenAI's own format documentation still describes only the single card. Our creative specs guide now says so.
Total budgets pace, and the euro minimum appears
Campaigns with a campaign-total budget now pace, spreading spend across the flight. The Help Center's Create Campaigns article, updated the same week, is explicit that the total budget is "the stricter control for total spend", that daily budgets are seven-day averages, and that the choice is only partly reversible: a total budget can become a daily one, and that campaign can never go back.
The same article carries the change nobody announced. On 2 September its Minimum Campaign Spend table listed ten billing currencies and no euro, and we wrote, on this site and in our guides, that any euro minimum was inference. On 7 September the table lists 22.
Currency | Minimum daily budget |
|---|---|
EUR | 15 |
GBP | 15 |
USD | 25 |
CHF | 20 |
CZK | 350 |
DKK | 110 |
HUF | 5,500 |
NOK | 160 |
PLN | 65 |
RON | 80 |
SEK | 175 |
QAR | 70 |
AED | 70 |
INR | 725 |
AUD, CAD, NZD | 25 |
BRL | 40 |
MXN | 150 |
JPY | 2,500 |
KRW | 25,000 |
15 EUR a day is about 450 EUR a month, the cheapest published entry after India's 725 INR, and it is now the floor we quote. Every page on this site that said "no published euro minimum" has been corrected.
The dirham row
One line in that table deserves its own paragraph. AED 70 is a minimum daily budget for accounts billed in UAE dirhams. OpenAI's availability table, re-read on 7 September, still lists 52 self-service countries and the United Arab Emirates is not among them; neither is Saudi Arabia. A billing row for a country with no availability row is a contradiction OpenAI has not explained. It could mean partner-managed UAE accounts exist through the Ads Solutions team, or that the availability table lags the billing system. Until a UAE-registered business can actually sign up, our UAE page stays as it is, and it now records the dirham.
Announced, and only announced
The email also says OpenAI "will be introducing" a conversion optimization model that counts view-through conversions as well as click conversions, billed on impressions. Today view-through conversions are supplemental reporting: they do not touch bidding, billing or the main Conversions column, and the Measure Results article says so. We will treat the new model as available when it appears in an account we run.
And the rollout line: "starting this week, we are rolling out ChatGPT Ads to select countries in Europe, India, the Middle East, and North Africa". That sentence describes 31 August. No country was added between then and 7 September. Where ChatGPT Ads are available has the current list and the dated changelog.
Sources
OpenAI, ChatGPT Ads product update email, 4 September 2026.
OpenAI Help Center, Create Campaigns for ChatGPT Ads, Minimum Campaign Spend table, read 7 September 2026.
OpenAI Help Center, Set up Custom Audiences for your Campaign, updated 5 September 2026.
OpenAI Help Center, Ads Manager Availability, read 7 September 2026.
developers.openai.com, Measurement Pixel reference.
Frequently asked questions
What is the ChatGPT Ads Manager plugin?+
A plugin for ChatGPT and Codex, announced by OpenAI on 4 September 2026, that creates, manages and analyses ChatGPT Ads campaigns from plain-language instructions. It turns a website or brief into ready-to-launch ads, creates variants, updates campaigns, troubleshoots delivery and proposes prioritised changes, previewing each one and asking for confirmation before applying it. It is installed from Plugins and connected to an advertiser account, and access depends on plan, workspace, role and region.
What is the minimum daily budget for ChatGPT Ads in euros?+
15 EUR a day, published on OpenAI's Create Campaigns article in the first week of September 2026, which is about 450 EUR a month. A week earlier the table had ten currencies and no euro row. It now has 22, including 70 QAR, 70 AED, 20 CHF and 175 SEK.
What changed for custom audiences on ChatGPT Ads?+
Members can be added, removed or replaced without recreating the audience; one upload can mix identifier types; audiences under 25,000 matched users can be used for exclusions; audiences can exceed five million members; size ranges are more granular; and Google Advertising ID is a supported identifier. Inclusion and bid multipliers still need 25,000 matched users.
Can a ChatGPT Ads campaign use a total budget?+
Yes. Campaigns take either a daily budget, a seven-day average, or a campaign-total budget, and since September 2026 total budgets pace so spend spreads across the flight. Changing a total budget to a daily one cannot be reversed for that campaign.
Is ChatGPT Ads available in the UAE now?+
No. OpenAI's availability table, read on 7 September 2026, lists 52 self-service countries and the United Arab Emirates is not one of them. The same week OpenAI published a 70 AED daily minimum, a billing row for a country that is not on the availability list. OpenAI has not explained the gap.
Which brands does ChatGPT name for your buyers' questions?
Explore with AI

Co-founder & Head of AI Innovation, Index Lab
Guilherme co-founded Index Lab, an AEO/GEO agency that makes brands the answer AI gives across ChatGPT, Gemini and Perplexity, taking clients from zero AI visibility to top recommendations.
[ Related reading ]
Listicles still dominate Google's top 10, but the winners keep turning over
New research on 3,000 ranking listicles finds the format never left. What changed is which URL wins, and what makes one listicle outrank another.
Read →NewsGoogle AI Mode picks different sellers than the shopping carousel
New tracking data finds AI Mode overlaps with Google's own product carousel only 1.28% of the time, and swaps in a different top seller for 49.6% of the matches it does find.
Read →NewsProduct pages beat Reddit for AI citations, but the data is B2B only
Ten Speed's study found product pages draw 24.1% of AI citations against 4.2% for Reddit, YouTube and forums combined, and the number survived six fact-check questions. The dataset behind it is B2B SaaS and professional services clients only.
Read →See where your brand stands in AI.
Free visibility check →