As OpenAI’s ChatGPT ad delivery improves, the doubts it created aren’t so easily fixed

OpenAI’s ChatGPT ad pilot is showing improving fill rates, up as much as 30% to 50% from launch levels over the last six weeks, but early underdelivery left many advertisers frustrated. One agency reportedly spent only $2,500 of a $250,000 commitment over four weeks, and several blue-chip advertisers are said to be reluctant to return without a trusted intermediary. OpenAI has since added CPC bidding, conversion tracking, and a pixel-based measurement API, but the article suggests the ad business remains in an early, trust-building phase.

Analysis

The key signal here is not near-term revenue for the new platform; it is validation that conversational ad units can produce search-like intent efficiency once supply normalizes. That has two implications: first, incremental spend is likely to come from innovation budgets before it migrates into core performance budgets; second, the real competitive threat is to long-dated search share, not social display, because the session context is now closer to high-intent discovery than passive feed consumption.

For Google, the market should focus less on immediate cannibalization and more on the option value of a new ad surface that validates the category. If conversational ads prove out, GOOGL can replicate the format across its own Gemini and search surfaces faster than a standalone AI company can mature a full ad stack, which makes this a medium-term moat expansion story rather than a binary loss. Meta is more insulated because its auction is still driven by identity and creative scale, but any reallocation of experimental budgets toward high-intent chat environments subtly reduces the share of wallet available for upper-funnel testing.

The open question is whether the current frustration creates a trust gap that takes quarters to repair. Advertisers will tolerate early inefficiency if they can see a path to measurement and spend control; they will not tolerate opaque underdelivery if procurement is involved. That makes the next catalyst less about ad volume and more about infrastructure milestones—self-serve, conversion APIs, and reliable budget release mechanics—because those determine whether budgets become repeatable demand or a one-off curiosity.

Contrarian view: the sell-side may be overindexing on current underdelivery as evidence of product weakness when it may actually indicate deliberate scarcity from a platform optimizing for user trust. If fill rates improve 30-50% from launch while performance remains near search benchmarks, the model can re-rate quickly on much smaller absolute spend than bulls expect. Netflix is the cleaner comparison operationally, but OpenAI’s path to advertising scale is more platform-like than media-like, so the market should price a slower start but a potentially steeper later ramp.

Key Decisions for Investors

• Add to GOOGL on any post-narrative weakness; 3-6 month horizon. If conversational ads mature, Google is the most likely beneficiary of category expansion rather than the main loser, with upside from faster internal product replication and a larger addressable ad surface.
• Buy META vs. short basket of smaller ad-tech beta names on a 2-4 month view. The market may over-rotate experimental budget shifts toward AI chat, but Meta’s performance engine and measurement stack should hold share better than ad-tech intermediaries.
• Hold off on chasing NFLX on this headline alone; if anything, use strength to sell covered calls 6-12 weeks out. NFLX is the closest precedent for early ad pain, but the current setup is too small to change the earnings trajectory meaningfully.
• For event-driven positioning, consider short-dated GOOGL calls only if there is a visible acceleration in conversational ad adoption or a partner/measurement announcement; otherwise avoid paying for a theme that is still infra-bound, not revenue-bound.
• Monitor for a reversal trigger: a credible self-serve and measurement rollout plus documented spend-release discipline. If those land within one quarter, rotate from defensive skepticism into a small basket long of AI monetization beneficiaries, led by GOOGL.