Here's How Much Revenue OpenAI's New Advertising Business Could Generate by 2030

OpenAI will begin testing labeled advertisements on ChatGPT’s free and low-cost tiers in the U.S., while excluding paid tiers (Plus, Pro, Business, Enterprise) and pledging not to sell user conversations or let ads drive responses. The company reported roughly $20 billion in revenue in 2025 and 900 million weekly users as of December 2025; Evercore ISI’s Mark Mahaney estimates the ad business could reach as much as $25 billion in annual revenue by 2030 if execution and user experience hold. The move could materially expand OpenAI’s monetization mix and poses a competitive threat to Google’s core commercial queries, though investor focus will center on potential user backlash and retention impacts.

Analysis

Market structure: OpenAI’s ad test turns ChatGPT into a high-intent, conversational ad inventory that could materially reallocate search/ad dollars; Evercore’s $25bn-by-2030 scenario is plausible if OpenAI converts even 5–10% of commercial search spend into first‑party ad revenue, which would be ~ $10–25bn and more than double OpenAI’s 2025 revenue base. Winners are AI infrastructure (higher inference volume), direct-response advertisers and performance-oriented ad sellers; losers are incumbents whose margins rely on search/query monetization (principally Alphabet). CPM/auction mechanics will shift toward quality and conversion metrics rather than click volume, lifting pricing power for platforms with superior measurement.

Risks: Tail risks include swift regulatory action (antitrust/privacy) that could ban certain ad targeting or force data segregation, and operational risks where bad ad integration or hallucinations cause MAU/engagement drops >10%, collapsing ad yield. Immediate (days–weeks): ad testing noise and short-term MAU volatility; short-term (3–12 months): advertiser onboarding and measurement proof points; long-term (2–5 years): structural reallocation of search/ad budgets. Hidden dependencies: advertiser ROI measurement, identity resolution, and OpenAI’s willingness to limit ad load to preserve Premium subscriber economics.

Trade implications: Tactical pair trades make sense — express conviction in AI infrastructure and hedge search-ad exposure. Use options to cap cost: 9–12 month put spreads on GOOGL to express downside if ad share erodes, and 12–36 month long exposure to NVDA (or AI infra ETFs) to capture higher inference demand. Rotate out of high-search-ad revenue names (reduce weight by 20–40% of ad-risk exposure) into AI infra, measurement vendors, and alternative ad channels over next 3–9 months as conversion metrics arrive.

Contrarian view: Consensus exaggerates near-term bleed from Google — historical shifts (e.g., social commerce) show incumbents retain pricing power via auction dynamics and measurement leverage. The market may underprice the difficulty advertisers face integrating ChatGPT conversions into funnels; if OpenAI fails to deliver reliably attributable ROAS within 12 months, ad CPMs will compress and adoption stalls. Watch two thresholds: weekly active users decline >5% post-ads launch or advertiser repeat-buy rate <50% after 6 months — both would signal strategy failure.

Key Decisions for Investors

• Establish a 1–1.5% portfolio short on GOOGL using a 12‑month put spread (buy 25% OTM, sell 40% OTM) to limit cost; target 10–25% downside over 6–18 months if OpenAI captures ad share—cut if Google quarterly ad revenue declines <3% QoQ or broader market turmoil increases.
• Initiate a 1.5–2% long in NVDA (or NVDA-dominant AI infra ETF) to capture increased inference demand; horizon 12–36 months, take profits at +30% or trim if order/intel shows OpenAI inference load growth <20% YoY.
• Execute a pair trade: long 1% ad-measurement/attribution vendor (e.g., The Trade Desk) and short 1% Alphabet (GOOGL) to express view that measurement winners gain share; re-balance after 2 quarterly advertiser-revenue print cycles.
• Reduce exposure to holdings where >30% revenue is search-ad dependent by 20–40% over the next 60 days; redeploy proceeds into AI infrastructure, ad-measurement, and privacy-compliant ID solutions.
• Set event-based monitors: if ChatGPT weekly active users fall >5% within 90 days post-ads OR advertiser repeat-buy rate <50% at 6 months, tighten stops on longs in AI infra by 10–20% and pause further ad-dependent short additions.