OpenAI's Financial Services Launch Isn't a New Beginning. It's a Catch-Up Move.

OpenAI launched ChatGPT for Financial Services on Thursday, combining its latest GPT-6 Astra model with institutional data to automate the work Wall Street analysts do before they get to make judgment calls — pulling data, building pitchbooks, reconciling financials, and formatting client deliverables. It was developed with design partners Morgan Stanley and Evercore, and bundles financial data from providers like PitchBook, Daloopa, and LSEG News.

There's just one problem with treating this as OpenAI's fresh offensive into finance. Anthropic launched a remarkably similar product four months ago.

Anthropic released 10 ready-to-run agent templates for financial services in May — covering pitchbook building, earnings review, financial modeling, KYC screening, month-end closing, and valuation review. They ship through Claude Cowork and Claude Code, integrate with enterprise approval flows, and are available immediately through Anthropic's financial services marketplace. OpenAI's product was announced today. Anthropic's was live four months ago.

This isn't about who built the better pitchbook assistant. It's about what each product reveals as both companies prepare to go public — and what that tells investors who haven't been able to buy a share of either one yet.

The enterprise race isn't about the product launch. It's about the numbers.

OpenAI's enterprise revenue finally exceeded consumer revenue for the first time in July 2026. That milestone matters because OpenAI's business model has been shifting from ChatGPT subscriptions — where it competes with itself for consumer attention — toward enterprise contracts, where it competes with Anthropic for Wall Street IT budgets.

But look at the underlying numbers. OpenAI reached approximately $40 billion in annualized revenue as of August 2026, up from roughly $25 billion through the spring. For the quarter that ended in June, it generated $6.7 billion in revenue, an 18% increase from the first quarter. That's strong growth.

Anthropic crossed $47 billion in annualized revenue in May — roughly double OpenAI's rate six months ago, and still higher today. Anthropic grew 80-fold in the first quarter of 2026, compared to OpenAI's three- to four-fold growth. The speed difference isn't a rounding error. It suggests enterprises have been choosing Anthropic's Claude models at a materially higher rate, particularly for the agentic workflows — like financial analysis — that require more than chat.

And the losses separate them further. OpenAI reported an operating loss of approximately $20.9 billion against $13.1 billion in booked 2025 revenue — losing roughly $1.22 for every dollar it earned. That burn is driven by the enormous compute costs of training new models and paying Microsoft for cloud capacity. Anthropic's burn rate is lower, and at $47 billion ARR it has more revenue to offset those same infrastructure costs.

Both companies are burning money to build the future. The question is which one can stop burning first.

What this product actually represents for OpenAI's economics

ChatGPT for Financial Services doesn't have a published price. It's sold through custom enterprise quotes — the seat count, usage allowances, data entitlements, and implementation costs all vary by firm. That's typical for enterprise deals, but it also means we can't tell how much revenue this specific product will add to OpenAI's bottom line.

What we can see is the pattern this represents. OpenAI is building verticalized products — finance, healthcare, education — to make enterprise sales easier. Instead of selling raw API access and letting a bank's IT team figure out compliance, data integration, and workflow design, OpenAI ships a finished solution with the data already bundled. The firm-specific templates for Excel, Word, and PowerPoint mean firm-specific Excel, Word, and PowerPoint templates mean Morgan Stanley's analysts get Morgan Stanley-formatted output without any custom integration work.

That's a real competitive advantage in sales velocity. A bank's procurement team can evaluate one packaged product instead of negotiating separate contracts with data providers, model vendors, and integration partners. OpenAI is trying to compress the sales cycle and lock firms into its ecosystem.

But Anthropic's approach — reference architectures that firms adapt to their own workflows, risk policies, and compliance standards — may be better suited to how financial institutions actually adopt new technology. Banks don't want a one-size-fits-all product. They want tools they can modify, govern, and audit. Anthropic's agent templates were designed around that preference from the start.

The IPO frame

Both companies confidentially filed their IPO prospectuses with the SEC in June 2026. OpenAI's CFO Sarah Friar told employees in August the company "will be a public company in 2027," though a strong growth trajectory could accelerate that timeline. Anthropic has been holding investor meetings and could debut as early as September.

OpenAI is valued at approximately $852 billion. Anthropic closed a $65 billion funding round in May at a $965 billion valuation — about $113 billion more. At those valuations, the market is pricing in years of future revenue from companies that are still deeply unprofitable.

The first one to go public sets the tone for the second. If the debut stocks strongly, it validates the entire generative AI enterprise thesis and lifts both companies. If it falters, the other faces pressure to delay or repricing. Both filed at roughly the same time precisely because neither wants to be left hanging while the other establishes the public-market benchmark.

For an investor watching this race, the financial services product launch is a useful signal — but not the one the headline suggests. The signal is that OpenAI's enterprise push is still playing catch-up. Its products are coming online, its enterprise revenue has crossed consumer revenue for the first time, and it's raising the valuation stakes ahead of the IPO. But Anthropic got to the same destination earlier, with more revenue, faster growth, and a narrower gap to profitability.

The question when these companies hit the public market won't be whether AI transforms financial services. That's already happening. It will be which company did the transforming first, and at what cost.