The Wall Street Journal published the report Tuesday morning. By the time the New York open hit, Oracle was down, Nvidia was lower, every name on the OpenAI capex chain was bleeding.

The numbers, per WSJ sourcing: OpenAI missed multiple monthly revenue targets in late 2025 and early 2026. ChatGPT failed to hit the company's internal goal of 1 billion weekly active users by year-end 2025. The shortfall in both cases was attributed to a single cause. Anthropic and Google's Gemini are taking share.

Reuters, CNBC, Forbes, Sherwood and the Decoder all matched within hours. Markets did the rest.

The Traffic Number That Should Scare OpenAI

Buried in the trading desk notes was the cleanest data point in the WSJ piece: ChatGPT's share of generative AI web traffic fell from 86.7% a year earlier to 64.5% in January 2026. Gemini's share went from 5.7% to 21.5% over the same period. That is not a wobble. That is a 22-point market share transfer in 12 months, in the largest consumer AI category, from the company that defined it to the company most analysts spent 2024 writing off.

Anthropic, for its part, did most of the damage in enterprise. Coding tools and agent harnesses are where Claude's revenue tripled this quarter, and where OpenAI's Codex ambitions have stalled in product chaos. Three senior executives left OpenAI on April 17, OpenAI for Science was dissolved, Sora was shut down at a reported $15 million per day in losses, and Prism was sunset three months after launch. The pattern is consolidation around Codex, the same week Anthropic ate OpenAI's developer base.

Wedbush, Gabelli, Piper Sandler all came out within hours saying the same thing in different words. "Largely a rehash of what we already knew," Gabelli's John Belton told CNBC. "OpenAI's growth seems to have slowed in late-2025 into early-2026 as the business ceded some share to Anthropic and Gemini." That is the bull-case framing. The bear case is harder.

The IPO Implication

Three things were supposed to support OpenAI's IPO valuation.

One: revenue growth at hyperscaler trajectory. Two: the implicit Microsoft monopoly on hosting and distribution. Three: ChatGPT as the consumer AI default.

All three lost altitude this week.

The Microsoft amendment announced Monday morning ended exclusivity through 2032. The WSJ scoop said the revenue trajectory is below internal targets. The traffic share number says ChatGPT is no longer the default, it is the leader of a three-way race that is going to be a two-way race by the end of the year. And every one of those three things now has to show up in the company's S-1 filing as a risk factor a federal court will hold them to.

OpenAI is reportedly targeting a public listing valuation in the $400 billion to $500 billion range. The secondary market is pricing it at $880 billion to $1 trillion. The gap, if you take the WSJ reporting at face value, just got harder to defend.

It Lands the Day Before Mag 7 Earnings

Wednesday, April 29: Microsoft, Alphabet, Meta and Amazon all report Q1 earnings after market close. Combined 2026 capex guidance going in was $115 billion to $135 billion at Meta, $175 billion to $185 billion at Alphabet, and roughly $200 billion at Amazon. The earnings calls were always going to be about whether AI investment is converting to revenue. After Tuesday morning, every analyst on those calls is going to ask one specific question, in eight different ways: are you taking share from OpenAI?

Microsoft will say yes, citing GitHub Copilot wins. Alphabet will say yes, citing Gemini's 21.5% traffic share, and Sundar Pichai will read the number twice. Meta will pivot the conversation to Muse Spark and capex efficiency. Amazon will cite Anthropic on Bedrock and Andy Jassy will note OpenAI is coming to AWS soon, an answer that is going to be quoted in OpenAI's own S-1 risk factors.

The OpenAI revenue miss did not just damage one private company on the eve of an IPO. It set the dialogue script for the most important earnings week of 2026.

The Bigger Question

OpenAI is still the most consequential AI company in the world. ChatGPT is still the largest consumer AI product, by a factor of three. The company is still profitable on a unit basis on its enterprise tier. None of that is in question.

What is in question is whether the trajectory was always going to be S-curved or whether the assumption inside the company was permanent linear scaling. The first month OpenAI missed an internal target was the month the assumption broke. The first quarter it missed multiple targets is the quarter the bankers running the IPO have to rewrite the pitch.

Anthropic is closing in. Gemini is the comeback story of the year. DeepSeek V4 just launched on Huawei chips, the markets did not even shudder. Open-weight models are eating the bottom of the pricing column. The middle of the market, where most of OpenAI's enterprise revenue lives, is where Anthropic is running the table.

The IPO was the victory lap. Now it is a recovery story. That is a different valuation.

Sources: Wall Street Journal (original scoop, April 28), Reuters (April 28), CNBC (April 28), Forbes (April 28), Sherwood (April 28), TradingView (April 28), the Decoder (April 28), 24/7 Wall Street (April 28), Yahoo Finance (April 28). Mag 7 earnings preview from CoinDCX, Investing.com, Wall Street Horizon, Blockonomi (April 28).