To absolutely no one’s surprise…

OpenAI has been dragging its feet on releasing ChatGPT-5 the past few weeks…

Because they wanted to see how xAI’s Grok 4 and Anthropic’s Claude 4 would perform before dropping ChatGPT-5.

They wanted to let the competition show their cards, then come in swinging.

Leaked ChatGPT-5 Performance Results Blew Grok 4 Out Of The Water

HLE stands for Humanity’s Last Exam.

Simply put, this benchmark is packed with the hardest math, logic, and reasoning questions that stumps even top human experts.

Most AI models, even the best from last year, struggled to break 20% on this test.

Grok 4 made headlines by hitting 44%, but ChatGPT-5 just raised the bar again.

To put it simply:

• A 56% score means ChatGPT-5 can solve problems at the level of a top college professor or PhD.
  

• Think of it as something that can help you solve PhD-level coding, business planning, or engineering problems instead of just writing basic emails.
  

• For business owners, this means you’ll have an AI that can break down complex data, solve tough problems, and give you answers you can trust better than Grok 4 or other LLMs.

Expected Release Date For ChatGPT-5

So, when do we get our hands on it?

OpenAI’s Sam Altman says “summer,” and most tech insiders are betting on a July or August launch. The official date is still under wraps, but all signs point to a release before the end of this quarter.

Talk Soon.

Brian

P.S.

I know I just talked last week about how amazing Grok 4 was, but the rate of AI development is just crazy that even I feel like I couldn’t catch a break making these announcements.