Sam Altman's “Last Resort” Just Became a $1 Billion Business. What It Means for OpenAI's IPO.

The important market signal is not the ad run-rate itself, but that OpenAI is moving from pure product growth to monetization that still leaves the business deeply cash-hungry. That keeps the medium-term demand case for compute intact, which is incrementally supportive for NVDA, but it also makes the eventual IPO harder to defend on a revenue-multiple basis unless margins improve sharply over the next 4-6 quarters.

For META, the near-term read is more defensive than bearish. OpenAI’s ad layer validates the direction of travel for AI-native advertising, but Meta still owns the superior auction stack, measurement loop, and advertiser trust, which are the real moats once performance budgets get reallocated. The bigger second-order loser is any premium valuation for private AI names that rely on consumer growth without showing durable monetization per user.

The main risks are trust and regulation, not competitive response. In the next 1-3 months, any sign that ad load is degrading engagement or that EU compliance slows rollout would compress the IPO narrative fast; over 6-18 months, the constraint is capital intensity, because every incremental monetization layer likely requires more inference spend. Contrarian view: consensus may be overestimating how easily a conversational interface can monetize without hurting usage quality, so the current enthusiasm could fade if advertiser ROI proves less stable than in search or social feeds.