#

Key takeaways

• Ramp data covers more than 70,000 U.S. businesses.

• Anthropic held nearly 44% of the segment's AI spending in July.

• OpenAI held nearly 40% and was growing faster in Q3 to date.

• The report says this is not a measure of the total market.

• Vendor choice should still depend on workload fit, price, retention, and governance.

What the Ramp data shows

TechCrunch reported on August 20, 2026 that Ramp data points to a close race between Anthropic and OpenAI among business users. The dataset covers more than 70,000 U.S. businesses. In July, Anthropic accounted for nearly 44% of the segment's AI spending, while OpenAI accounted for nearly 40%.

The report also says OpenAI was growing faster in Q3 to date. That matters because it suggests momentum can shift even when one vendor still leads in a given period. The numbers are close enough that small changes in spending could alter the picture.

What the report does and does not measure

The report includes an important caution. It is not a measure of the total market. That means the figures should be read as a view into one spending segment, not as a full industry ranking.

This distinction matters for any reader trying to use the data as a proxy for broader demand. A narrow dataset can show useful trends, but it cannot answer every market question. The safest reading is that the data reflects one slice of business spending.

How to interpret the spending split

The July split shows two vendors with similar shares. Anthropic leads in the reported segment, but OpenAI is close behind. OpenAI's faster growth in Q3 to date adds another layer to the comparison.

That combination can point to changing buyer preferences, but the report does not explain why the shift is happening. It also does not say whether the trend will continue. Readers should treat the data as a snapshot, not a forecast.

Operational considerations for buyers

The report itself points to the right decision criteria. Buyers should evaluate AI vendors against workload fit, price, retention, and governance. Those factors are more useful than a single market-share signal.

Workload fit asks whether the tool matches the task. Price affects adoption and scale. Retention and governance matter because enterprise use depends on keeping systems manageable and controlled.

Morocco relevance

The source reports no Morocco-specific facts. For readers in Morocco, the global lesson is to avoid treating market-share headlines as local availability or advice. Use the same vendor checks the report highlights: workload fit, price, retention, and governance.

Why this matters for enterprise AI decisions

Spending data can be helpful, but it should not drive decisions on its own. A vendor with rising spend may still be the wrong fit for a specific team. A vendor with a smaller share may still be the better operational choice.

That is why the report's caution is important. It separates market attention from practical selection. For enterprise buyers, the best decision process stays grounded in use case needs and internal controls.

Bottom line

Ramp's data suggests a tight contest between Anthropic and OpenAI in one business spending segment. OpenAI is growing faster in Q3 to date, but Anthropic still leads in July share. The report does not claim to show the full market.

The clearest takeaway is simple. Use market signals as context, not as a substitute for vendor evaluation. The report's own criteria remain the most relevant ones: workload fit, price, retention, and governance.