Ben-R-F week for OpenAI, facing so much user backlash over its highly promoted GPT-5 rollout, had to bring back a previous model, but OpenAI might be making some new inroads in a more lucrative part of the business. McKenzie Segalo says that in today's tech check. Morning, Mac. Hey, good morning, Carl. It was specifically those chatbot users that were put off by the so-called new personality of

GBT5.

They said that it felt colder and less intuitive.

But while they were loudly complaining, enterprise customers have quietly been leaning in.

And it's that business opportunity which OpenAI is aggressively courting with this upgrade.

Now, these are deep pocketed and sticky buyers that sign up to multi-year spending deals.

And I've been talking to startups out here in Silicon Valley, many of them long-time

anthropic customers.

And they're now making GBT5 the default in certain products and workflows, especially for

tasks like code generation, debugging, and

design. As one CTO put it, it's just more creative than Anthropics models. Founders are

also telling me that GBT5 is finally doing what older models couldn't, best in class coding

that gives reliable and consistent results, and that matters a lot when you're building apps.

It's also showing up in OpenAI's internal adoption numbers.

So developers building on OpenAI's tech have more than doubled since launching.

Coding and agent building work has spiked, and people using GBT5 for more sophisticated reasoning use cases

are up eightfold and that is big news because this is where Anthropic has been a market leader.

80% of its business comes from these enterprise customers so a shift in sentiment here is key.

OpenAI already works with the likes of JPMorgan, T-Mobile, and SoftBank

on consumer chatbot solutions, but this is a new kind of win. GBT was not known for

strong coding performance before, and at this point, it's beginning to gain traction, and that's a

major unlock for the kind of customer that Anthropic has long dominated.

...