‘Maybe ChatGPT is a different kind of a place’: Why OpenAI is saying no to search budgets — for now

OpenAI is testing ChatGPT ads with CPC-based buying, but is explicitly steering advertisers away from search budgets for now and toward test budgets and upper-funnel spend. Conversion tracking has only just launched, CPA bidding is next, and OpenAI has also lowered the minimum spend requirement to $50,000 to attract smaller advertisers. The article suggests the monetization strategy is still early-stage and constrained by limited measurement infrastructure.

Analysis

The market is still underestimating the sequencing here: OpenAI is intentionally monetizing attention before it can credibly monetize outcomes. That usually advantages the infrastructure provider in the short run, but it also means the first dollars are likely to come from experimental display/programmatic budgets rather than high-velocity search budgets, limiting near-term dollar volume while giving OpenAI time to tune relevance and pricing. In other words, this is a supply-constrained ad product disguised as a demand launch, which is why the revenue slope is likely shallow for several quarters even if usage grows quickly.

For Meta, the main risk is not immediate budget share loss, but a slower erosion of marginal testing spend and a shift in marketer learning behavior. If ChatGPT becomes a legitimate upper-funnel discovery surface, it can intercept the “cheap curiosity” dollars that historically incubate new products on Meta before conversion budgets follow. That is a second-order negative for Meta because it pressures the ecosystem that feeds its auction with fresh demand, especially among smaller advertisers who are easiest to reallocate and most sensitive to novelty.

CRTO is the cleaner beneficiary because its commercial position is leverage to OpenAI’s eventual need for performance plumbing. The key insight is that the current CPC framing is a bridge product: it trains advertisers and validates intent inventory while preserving user trust, but it also sets up a later migration to CPA and measurement tools where ad tech partners with identity, attribution, and dynamic optimization can matter more. If conversion tracking improves over the next 3-9 months, CRTO’s role could expand faster than the market expects, since the real bottleneck is not ad load but the ability to prove incrementality.

The contrarian view is that the consensus may be too focused on “search share stolen” and not enough on “new budget created.” If ChatGPT proves it can generate commercial queries without degrading trust, this could become an incremental category rather than a substitution trade, which would be bullish for ad-tech intermediaries more than for direct-search incumbents. The tail risk is reputational: any noticeable relevance decay or ad clutter could trigger a fast trust reset, and because this product is still in test mode, that would likely show up within weeks rather than years.

Key Decisions for Investors

• Long CRTO / short META pair for 1-3 months: express the view that ad-tech enabling the new monetization layer benefits faster than incumbent auction platforms lose budget; target 10-15% relative outperformance if OpenAI conversion/CPA messaging gains traction.
• Buy CRTO on 2-4 week pullbacks ahead of measurement/CPA rollout headlines; risk-reward favors upside because the market can re-rate the partner role before revenue inflects, while downside is capped if OpenAI adoption remains test-budget only.
• Short META tactically on any strength tied to AI ad monetization headlines, but keep size modest and use a 6-8 week horizon; thesis is budget cannibalization at the margin, not an immediate earnings reset, so this is a relative-value trade rather than a structural short.
• Sell out-of-the-money META puts 2-3 months out only if implied volatility remains elevated; the trade monetizes the market’s fear of AI-ad disruption while recognizing that near-term financial impact is likely small.
• If you want convexity, buy a CRTO call spread 3-6 months out financed by selling near-term upside in META; this captures the lag between product launch rhetoric and actual performance-budget migration.