ChatGPT Ads Complete Guide: What It Is, How It Works, and Whether You Should Be on It
ChatGPT Ads is not Google Search with a chatbot wrapper. It's not Meta with AI targeting. It's a third thing — a recommendation engine embedded in decision-making conversations. Understanding that difference is the difference between profitable early adoption and expensive confusion.
1. What ChatGPT Ads Actually Is
In February 2026, OpenAI began placing sponsored recommendations at the bottom of ChatGPT's answers. The surface-level story is simple: AI chatbot gets ads. The real story is more interesting.
ChatGPT Ads is the first advertising channel native to conversational AI. Every ad channel before it — search, social, display, video — was built around one of two human behaviors: looking for something (search intent) or killing time (attention). ChatGPT Ads sits on a third behavior: thinking through a decision with help.
Core insight: You're not bidding on keywords or interests. You're bidding on decision contexts.
When someone types "best CRM for a 5-person team" into Google, the search engine matches keywords to ads. When someone tells ChatGPT "we're a five-person team and our sales process is a mess, what should we use?" — the system matches the situation to a recommendation. The difference sounds subtle. It is not. One is query matching. The other is context matching. The data that feeds each is fundamentally different.
This is why every comparison to Google Ads or Meta Ads misses the point. ChatGPT Ads is not a cheaper version of search or a better-targeted version of social. It is a new category. And like most new categories, the people who understand it first capture disproportionate returns before the market catches up.
The Three-Way Comparison
| Google Search Ads | Meta Ads | ChatGPT Ads | |
|---|---|---|---|
| User state | Looking for something specific | Browsing, passing time | Thinking through a decision |
| Matching logic | Keywords + landing page relevance | Behavioral profile + engagement history | Conversation context + past chats |
| Advertiser's lever | Keyword bidding + ad copy | Creative + audience targeting | Context relevance + offer strength |
| User sees ad when | They type a query | They scroll through feed | They're mid-conversation about a related topic |
| Post-click behavior | User leaves Google | User leaves Meta | User can keep chatting — including with the ad |
2. The Timeline: How "Last Resort" Became a $100 Billion Ambition in 18 Months
The timeline matters not for the dates, but for what it reveals about the velocity and the pressure driving this platform.
2024, May: Sam Altman says "ads + AI makes me uncomfortable. It would be a last resort."
2025, December: The Information reports that OpenAI employees are designing "sponsored content prioritization" — a system where paid recommendations can appear above organic ones.
2026, January 16: OpenAI officially announces ad testing in ChatGPT for US Free and Go ($8/month) users. Publishes five "ad principles": mission alignment, answer independence, conversation privacy, user choice and control, long-term value.
2026, February 9: Pilot launches. CPM-only. $60 CPM. Minimum spend north of $200,000. A dozen or so handpicked advertisers. Real-time reporting is unavailable — advertisers get aggregated data only.
2026, March 21: OpenAI tells Reuters ads will be shown to all US Free and Go users "in the coming weeks." Hires Dave Dugan, former Meta ad executive, to head ad sales.
2026, March 26: Pilot expands to Canada, Australia, and New Zealand. OpenAI says "no impact on consumer trust metrics, low dismissal rates."
2026, April: CPC bidding appears — $3 to $5 per click. Minimum spend drops to $50,000. The $60 CPM that launched in February is now clearing at $25 in some categories. Ten weeks from premium-priced experiment to market-price reality.
2026, May 5: Ads Manager Beta opens to all US businesses. No minimum spend. Self-serve. Conversions API and pixel measurement available.
2026, May 7: Expansion announced to UK, Mexico, Brazil, Japan, South Korea.
2026, June: All five new markets go live. Ads Manager gains daily budgets, CPM-to-CPC campaign cloning, custom bid caps, bulk editing, conversion tracking, and UTM parameters.
2026, June 17: OpenAI publishes Ad Tools Terms defining Audience Tools (first-party data upload for custom audiences) and Creative Tools (AI-generated, modified, and localized ad creative).
2026, July: UK Ads Manager opens. Japan and South Korea ads go live. First-party audience targeting rolls out to advertisers.
What this timeline actually says: OpenAI went from "we don't want ads" to a nine-market, self-serve, CPC-enabled, audience-targeted ad platform in under six months of public operation. That is not the pace of a company cautiously experimenting. That is the pace of a company that has decided advertising is existential.
3. How the Ad Engine Works Under the Hood
Most coverage of ChatGPT Ads focuses on what users see: a sponsored label, a product recommendation, a link. That's the surface layer. The mechanism underneath is what determines whether your ads work.
3.1 Matching: It's Not Keywords. It's Conversation Context.
Google matches ads to queries by looking at the words you typed. The matching is lexical — "running shoes" triggers ads for running shoes. The system knows what you asked. It does not know why you asked it.
ChatGPT matches ads to conversations. The system sees the full context: what you've been discussing, what you asked before, what you clarified, what constraints you mentioned. When someone says "I need running shoes but my knees hurt after 5K," the ad engine knows this is about cushioned stability shoes for a runner with joint pain — not just "running shoes."
The practical implication: Keyword-based ad optimization, the core competency of search marketers for 20 years, does not transfer. What transfers is understanding the decision journey your customer is on and positioning your offer at the right moment in that journey.
OpenAI has stated that ads are matched based on:
- The topic of the current conversation
- Past chats and conversation history
- Past interactions with ads (which ads the user engaged with or dismissed)
The system does not share conversation content with advertisers. Advertisers receive only aggregated performance data.
3.2 The Auction: CPM → CPC → Conversions
Every ad platform follows a predictable evolution:
- CPM phase: Platform sells impressions, advertiser bears the performance risk. Platform proves reach.
- CPC phase: Platform shares performance risk. Advertiser only pays when someone clicks. Platform proves engagement.
- Conversions phase: Platform optimizes for outcomes. Advertiser pays for results. Platform proves ROI.
ChatGPT Ads launched at stage 1 in February, moved to stage 2 by April, and is now in the early innings of stage 3. Ad Age reported on July 22 that OpenAI is "making the transition to conversions, proving its place in the ad channel hierarchy by generating proven sales results."
This transition is the single most important thing to watch. A platform that can't prove it drives sales will always be a "test budget" line item. A platform that can prove it drives sales becomes a core channel. Google went through this transition around 2004-2006. Meta went through it around 2013-2015. ChatGPT is going through it right now.
3.3 Attribution: What You Can and Cannot Measure
As of July 2026, ChatGPT Ads offers:
| Available | Not Yet Available |
|---|---|
| Impressions, clicks, CTR | View-through conversions |
| Avg CPC, Avg CPM | Multi-touch attribution |
| Conversions (via pixel + CAPI) | Incrementality testing tools |
| UTM parameters for traffic | Cross-device attribution |
| First-party audience matching | Audience overlap and frequency reports |
The conversions infrastructure exists. Whether it works well enough for advertisers to trust their budgets to it is the open question.
In April 2026, Digiday reported that "OpenAI's first advertisers can't prove ChatGPT Ads work." That was three months ago. Since then, Conversions API, pixel measurement, and UTM support have all shipped. The tooling is improving. But the burden of proof is still on the platform.
3.4 Where Ads Appear (and Where They Don't)
Ads appear at the bottom of ChatGPT's answers, clearly labeled as "Sponsored" and visually separated from the organic response. They are shown only to:
- Logged-in adult users
- On the Free tier or the $8/month Go tier
They are not shown to:
- Plus, Pro, Business, Enterprise, or Education subscribers
- Users under 18
- Conversations near sensitive topics (health, mental health, politics)
Free-tier users can opt out of ads — but doing so reduces their daily message allowance. This creates a clean tier structure: pay with money (subscription), pay with attention (ads), or pay with both (Go tier).
4. Pricing: The Numbers and What They Signal
4.1 Current Pricing
| Metric | Range | Notes |
|---|---|---|
| CPM | $25–$60 | Launched at $60 fixed; now $25+ with custom bid caps |
| CPC | $3–$5 | Mid-range; comparable to mid-funnel Google Search in competitive categories |
| Minimum spend | $0 | Self-serve Ads Manager has no minimum; earlier managed-service pilots required $50K–$250K |
4.2 Pricing by Vertical (Approximate CPC Ranges)
| Vertical | CPC Range |
|---|---|
| SaaS / Technology | $3.00–$4.50 |
| E-commerce / Retail | $3.00–$5.00 |
| Local Services | $2.50–$4.00 |
| Travel & Experiences | $3.00–$5.00 |
These are early-market prices. As more advertisers enter and the auction matures, expect compression in some categories and inflation in others — same dynamic as every ad platform in history.
4.3 What the Pricing Collapse Tells Us
The $60 CPM that launched in February didn't hold for ten weeks. It dropped to $25 in some categories. Then CPC bidding arrived.
This is not a sign of failure. It's a sign of price discovery. Every new ad platform overprices at launch. The early adopters are high-intent brands willing to pay a premium for first-mover advantage. As the platform opens up, the market finds the clearing price where supply (ad inventory) meets demand (advertiser budgets).
The more interesting signal: OpenAI dropped prices and opened self-serve faster than almost any ad platform in memory. Google kept search ads on a managed-service model for years. Meta kept Instagram ads invitation-only for a long time. TikTok took its time. OpenAI went from $200K-minimum managed service to $0-minimum self-serve in about three months. That speed suggests one of two things: either the managed-service pipeline wasn't filling up, or OpenAI is prioritizing market share over short-term pricing power. Probably both.
5. Who Should Be on ChatGPT Ads Right Now — and Who Shouldn't
This is the question most marketers are actually asking. Here's a decision framework.
Likely Worth Testing Now
E-commerce brands with clear category alignment. If your product is the kind of thing people research, compare, and decide on conversationally — meal kits, skincare, travel gear, SaaS tools, furniture — the context-matching mechanism works in your favor. When someone tells ChatGPT "I'm planning a camping trip and need gear for three people," your camping gear ad isn't interrupting. It's relevant.
SaaS companies with a well-defined ICP. The detailed, conversational nature of ChatGPT queries means B2B buying signals are richer than in search. A user describing their team size, budget, pain points, and current stack is giving you more signal than any keyword could.
Brands that can afford to learn. This is not a performance channel yet. If you need every dollar to show attributable ROAS within 30 days, wait. If you can allocate $1,000–$3,000/month as a learning budget to understand a new channel before it matures, the window is open.
Probably Wait
Local service businesses without strong digital conversion paths. ChatGPT Ads can drive interest, but if your conversion path is "call for a quote" and you can't track which calls came from ChatGPT, you're flying blind.
Brands that rely on impulse purchase behavior. ChatGPT conversations are intentional. Users are thinking, not browsing. The impulse-buy dynamic that powers a lot of social commerce doesn't apply here. If your product sells because someone saw it, not because they researched it, this channel is not for you.
Anyone who can't deploy proper conversion tracking. If you can't implement the Conversions API or at minimum configure UTM + GA4 to track ChatGPT-sourced traffic through to conversion, you don't have a feedback loop. Without a feedback loop, you can't optimize. Without optimization, this is just spending money to learn nothing.
A Common Trap: Treating This Like Google Search
The trap: you take your Google Search campaigns, copy the keyword themes, and launch on ChatGPT Ads.
Why it fails: ChatGPT Ads does not match on keywords. It matches on conversation context. Your Google campaign targeting "best CRM for small business" assumes a specific query. On ChatGPT, the same intent might surface across dozens of conversation paths — "we're scaling and our spreadsheets are breaking," "what do you use for pipeline management?", "how do I stop losing track of follow-ups?" None of those would trigger your Google ad. All of them could trigger a ChatGPT recommendation.
The right approach: Instead of mapping keywords to ads, map decision moments to offers. What situation is your customer in when they're most likely to need what you sell? What are they actually saying to the AI in that moment? Optimize for those contexts, not those queries.
6. The Gaps: What's Still Missing and Why It Matters
ChatGPT Ads has built an impressive amount of infrastructure in six months. It's also missing capabilities that advertisers on mature platforms take for granted. Here are the gaps that matter.
6.1 Conversion Optimization at Scale
You can track conversions. You cannot yet tell the algorithm "optimize for conversions." On Google or Meta, you set a conversion event (purchase, lead, signup), set a target cost per action, and the algorithm automatically adjusts bids, placements, and audience matching to hit that target. ChatGPT Ads doesn't do this yet. You can bid CPC or CPM. You cannot bid CPA.
Why it matters: Manual optimization doesn't scale. If you can't automate toward outcomes, you can't deploy significant budgets efficiently. This is the single biggest technical gap.
6.2 Remarketing
You cannot retarget website visitors or past ad engagers within ChatGPT Ads. No pixel-based retargeting. No engagement-based retargeting. No list-based retargeting (though first-party audience upload for initial targeting is now available).
Why it matters: Remarketing is typically the highest-ROI segment in any ad account. Its absence means the funnel is incomplete — you can acquire interest, but you can't recapture it efficiently.
6.3 Lookalike / Similar Audiences
You can upload first-party customer lists to create custom audiences. You cannot yet tell the platform "find more people like my best customers."
Why it matters: Lookalike audiences are how you scale beyond your known customer base without losing efficiency. Without them, expansion campaigns are less targeted and typically less profitable.
6.4 A/B Testing Infrastructure
The Ads Manager does not have a structured creative testing framework. No split-test campaigns. No statistical significance indicators. No automated winner-declaration.
Why it matters: Creative testing is how you improve. Without it, you're guessing.
6.5 Rich Media Ad Formats
Ads are currently text + static image. No video. No carousel. No interactive formats (though OpenAI has shown concept images of interactive ads where users can "ask the ad questions").
Why it matters: Text + static image severely limits the types of products and messages that can be communicated. A furniture brand that sells on visual appeal, a SaaS company that demos on video, a travel company that sells on aspirational imagery — all are working with one hand tied behind their back.
6.6 Audience Insights and Reporting
Advertisers get performance metrics on their own campaigns. They do not get audience-level reporting — demographics, interests, behaviors — that helps them understand who is engaging and why.
Why it matters: Without audience insights, you can't feed learnings back into your broader marketing strategy. ChatGPT Ads is a black box in, a black box out.
7. Where This Is Heading: The AI-Native Ad Stack
The gaps above are not permanent. They're a snapshot of an immature platform. The direction of travel is clear, and it points toward something that doesn't exist yet: an AI-native advertising stack.
7.1 The Three Phases of AI Advertising
Phase 1 — AI as Ad Surface (Now): Ads are placed inside AI products. The AI is the container, not the mechanism. This is where we are today. ChatGPT shows ads the way a news website shows ads — alongside the content, not integrated into it.
Phase 2 — AI as Ad Optimizer (2027-2028): The AI itself begins optimizing ad delivery, creative, and targeting. OpenAI has already published terms for "Creative Tools" that generate, modify, and localize ad creative using its models. Audience Tools upload first-party data. The logical next step is the AI building and testing creative variants, adjusting bids, and refining targeting — all without human intervention.
Phase 3 — AI as Ad Destination (2029+): The user never leaves the AI. They ask about a product. The AI recommends it. The user asks follow-up questions. The AI answers them — pulling from the advertiser's product data, reviews, and specifications. The user buys. The transaction happens within the AI interface. The AI is not just the ad platform. It's the store.
This third phase is where OpenAI's $100 billion revenue ambition lives. Not in showing ads alongside answers, but in becoming the interface layer for all of commerce.
7.2 What This Means for Existing Ad Platforms
Google and Meta are not standing still. Google is building AI Mode in search — an answer-first experience that will inevitably include ads. Meta is embedding AI assistants across its apps. But both face the same structural challenge: their current ad businesses are built on platforms (search results pages, social feeds) that AI interfaces are designed to replace. The innovator's dilemma is real.
The platform that wins AI advertising won't be the one with the best ad tech. It'll be the one with the most trusted AI assistant. People don't ask Google for advice. They ask it for links. People do ask ChatGPT for advice. That trust — if it survives the introduction of advertising — is the moat.
7.3 The Competitive Landscape
| Platform | Ad Status (July 2026) | Strategic Position |
|---|---|---|
| ChatGPT | Live in 9 markets, self-serve open in US and UK | First mover. Fastest iteration. Most to lose if trust erodes. |
| Google Gemini | No ads. Google publicly denied 2026 launch plans. | Most to lose from AI advertising — it cannibalizes search ads. Will enter when forced, not when ready. |
| Claude (Anthropic) | "Claude will remain ad-free." Ran anti-ad Super Bowl campaign. | The ad-free alternative. A viable escape hatch if ChatGPT ads annoy users. |
| Perplexity | Exploring "sponsored answers." No self-serve platform. | Closest competitor. Positioned as "answer engine," naturally ad-friendly. Watching ChatGPT closely. |
| Meta AI | No standalone ads. Integrated into Meta's existing ad ecosystem. | Will almost certainly build AI-native ad formats. Distribution advantage through existing advertiser relationships. |
The most interesting dynamic: ChatGPT's AI traffic share has declined from roughly 87% to 65% as Gemini gains ground. If ads degrade the ChatGPT experience, the switching cost for users is near zero — there are multiple free alternatives. This is the tightrope OpenAI is walking.
8. The Revenue Ambition and the Reality Gap
OpenAI has told investors it expects ChatGPT Ads to generate:
| Year | Ad Revenue Target |
|---|---|
| 2026 | $2.5 billion |
| 2027 | $11 billion |
| 2028 | $25 billion |
| 2029 | $53 billion |
| 2030 | $100 billion |
These numbers assume OpenAI reaches 2.75 billion weekly users by 2030. For context, Google has approximately 4 billion+ weekly users across all products. OpenAI's target assumes it will reach roughly two-thirds of Google's current user base in four years, then monetize each user at rates competitive with search advertising.
Third-party analysts are skeptical. eMarketer projects the entire US chatbot advertising market will generate less than $1 billion in 2026 — meaning OpenAI alone would need to be 2.5x larger than the total market. By 2030, eMarketer forecasts the entire US AI chatbot ad market at $5.41 billion. OpenAI's $100 billion target is roughly 20 times that.
Somewhere between these two numbers lies reality. The key variables:
- How fast ChatGPT's user base grows (and whether ads slow that growth)
- Whether conversion attribution matures enough to unlock performance budgets
- Whether Google and Meta enter aggressively, fragmenting the market
- Whether users accept or reject AI advertising
None of this changes the near-term opportunity. Even if OpenAI hits 10% of its 2026 target — $250 million — that's still the fastest ramp of any new ad platform in a decade. And the advertisers who understand the channel before the performance tools mature will have an asymmetric advantage when they arrive.
9. What to Do Now: A Practical Checklist
If You Want to Be an Early Mover
Register an Ads Manager account. Go to ads.openai.com. Complete account setup and verification. Even if you don't spend a dollar, you'll see the interface, understand the campaign structure, and be ready when you want to move.
Audit your brand's presence in ChatGPT. Ask ChatGPT questions your customers ask. Does your brand appear in the answers? Are competitors being mentioned? Is your category being described accurately? This is your baseline. It's also your GEO (generative engine optimization) opportunity — being recommended organically and being visible in the ad slot are complementary, not competing.
Define your decision moments. Don't think about keywords. Think about the situations where your product becomes the obvious answer. "We're a team of five and our onboarding is manual and error-prone" is a decision moment. "What's a good HR tool?" is a query. ChatGPT Ads rewards the first, not the second.
Prepare your first-party data. Audience Tools allow you to upload customer lists for targeting. If you have clean customer data with emails or phone numbers, you're ahead of advertisers who don't. Organize it now.
Set a learning budget. $1,000–$3,000/month for 2-3 months. The goal is not ROAS. The goal is data. What contexts trigger your ads? What's the CTR? What's the downstream conversion behavior? You're buying information, not sales.
If You're Waiting
Monitor the conversion attribution story. When credible case studies emerge showing clear conversion paths from ChatGPT Ads to revenue — not just clicks, not just traffic, but attributable sales — that's your signal that the platform has crossed the threshold from experiment to channel.
Watch the pricing trend. If CPCs start compressing while conversion rates improve, early movers are getting better deals. If CPCs inflate without conversion data to justify them, the platform isn't ready and you're right to wait.
Track user sentiment. Reddit, X, and app store reviews are your leading indicators. If users start complaining about ads degrading the ChatGPT experience, retention may suffer and the platform's value proposition weakens. If users seem indifferent, the platform has room to grow.
10. The Bottom Line
ChatGPT Ads is the first genuine ad channel built for the AI era. It is immature, unproven at scale, and surrounded by financial projections that strain credibility. It is also, for the right advertiser with the right expectations, the most interesting new advertising surface since mobile.
The advertisers who win here will be the ones who understand that ChatGPT Ads is not a search channel or a social channel. It's a decision channel. The user is not looking for information or passing time. They are trying to make a choice. If your ad helps them make that choice — not interrupts them, not distracts them, but genuinely moves their decision forward — you win.
If your ad treats this like any other impression, you lose. The context is too rich and the user's state is too intentional for lazy advertising to work. That's not a limitation. It's a filter. And for now, before the platform matures and the auction gets crowded, the filter is on your side.
References
- OpenAI, "Testing ads in ChatGPT," May 7, 2026. https://openai.com/index/testing-ads-in-chatgpt/
- OpenAI, "Our approach to advertising and expanding access to ChatGPT," January 16, 2026. https://openai.com/index/our-approach-to-advertising-and-expanding-access
- OpenAI, "New ways to buy ChatGPT ads," May 5, 2026. https://openai.com/index/new-ways-to-buy-chatgpt-ads
- OpenAI Ads Manager. https://ads.openai.com
- Garett Sloane, "How ChatGPT advertising works now — and what's still missing," Ad Age, July 22, 2026. https://adage.com/technology/ai/aa-chatgpt-ads-ad-tech-what-marketers-still-need/
- Anisha Sircar, "OpenAI Brings Ads To ChatGPT As Costs Mount," Forbes, January 20, 2026. https://www.forbes.com/sites/anishasircar/2026/01/20/openai-brings-ads-to-chatgpt-as-costs-mount
- CNBC, "OpenAI to begin testing ads on ChatGPT in the U.S.," January 16, 2026. https://www.cnbc.com/2026/01/16/open-ai-chatgpt-ads-us.html
- Axios, "OpenAI launches self-serve ad platform," May 5, 2026. https://www.axios.com/2026/05/05/openai-self-serve-ad-platform
- The Next Web, "OpenAI's ChatGPT ads just went cost-per-click," April 2026. https://thenextweb.com/news/openai-chatgpt-cpc-ads-launch
- Search Engine Journal, "ChatGPT Ads Now Offer CPC Bidding Between $3 And $5: Report," April 2026. https://www.searchenginejournal.com/chatgpt-ads-now-offer-cpc-bidding-between-3-and-5-report/572652
- WASK, "How Much Do OpenAI Ad Cost in 2026? Pricing & CPM Guide." https://blog.wask.co/digital-marketing/openai-ad-pricing
- WebFX, "How Much Do ChatGPT Ads Cost? 2026 CPC & CPM Pricing." https://www.webfx.com/blog/ai/chatgpt-ads-cost
- Investing.com, "OpenAI projects $2.5 billion in ad revenue for 2026." https://www.investing.com/news/company-news/openai-projects-25-billion-in-ad-revenue-for-2026-93CH-4605384
- Yahoo Finance / eMarketer, "OpenAI Is on Pace to Miss Its Own Ad Revenue Forecast by 90%," July 2026. https://finance.yahoo.com/technology/ai/articles/openai-pace-miss-own-ad-142128219.html
- MediaPost, "OpenAI Ad Revenue Is Far Under Its Goal, Analysis States," July 21, 2026. https://www.mediapost.com/publications/article/416676/openai-ad-revenue-is-far-under-its-goal-analysis.html
- Bram Van der Hallen, "June 2026 Updates — ChatGPT Ads," LinkedIn. https://www.linkedin.com/pulse/june-2026-updates-chatgpt-advertising-bram-van-der-hallen-of00e
- Benji Shomair, "ChatGPT ads pilot expansion to Japan and South Korea," LinkedIn, July 2026. https://www.linkedin.com/posts/bshomair_were-continuing-the-global-expansion-of-activity-7474748028994732032-qEsn
- Digital Applied, "Ads in ChatGPT: OpenAI's New Advertiser Toolkit 2026," June 2026. https://www.digitalapplied.com/blog/openai-chatgpt-ads-audience-creative-tools-2026-guide
- Marketing Dive, "OpenAI solidifies ad platform ambitions with ChatGPT Ads Manager," May 2026. https://www.marketingdive.com/news/openai-solidifies-ad-platform-ambitions-with-chatgpt-ads-manager/819801
- Jay Lin, "ChatGPT廣告來臨，商家必知的行銷新機會與應對策略." https://imjaylin.com/chatgpt-ads
- Writesonic (@WriteSonic), X post on pilot revenue run-rate. https://x.com/WriteSonic/status/2047320572372406424
- MG Siegler (@mgsiegler), X post referencing Adweek revenue projections. https://x.com/mgsiegler/status/2076954801703625088
- The Information (@theinformation), X post on OpenAI lowering ad minimums. https://x.com/theinformation/status/2059265889527116200
- Flyweel, "ChatGPT Ads Manager: Self-Serve Launch Guide." https://www.flyweel.co/blog/openai-launches-chatgpt-ads
- Search Engine Land, "ChatGPT with ads: 'Free-user monetization' coming in 2026?" https://x.com/sengineland/status/1915453014397034516
Insight Artifact
让 Aura 为下一个问题产出 Artifact。
Aura 会调研市场、处理你授权的数据，并交付可分享、归档和持续复用的完整成果。