GPT-5.3-Codex vs Grok 4.5 for humanising AI text

The verdict

GPT-5.3-Codex wins, 9790

Same 12 humanising AI text tasks, marked blind by three rival labs. GPT-5.3-Codex took 5 tasks, Grok 4.5 took 1, 6 tied. Tested 16 Aug 2026.

Where they differed most

The task — Best man, not best template

This best-man speech opener could be about anyone: 'They say the definition of a best friend is someone who knows all your stories — well, ladies and gentlemen, I know all of Mark's, and after tonight, so will you. Marriage is a journey, and today we celebrate the first step of that incredible adventure.' Rewrite it around the two real details available — Mark once missed his own engagement party because he was up a ladder rescuing a cat, and he met the bride, Priya, when she reported him to the council (he was the noise complaint) — so that no other best man on earth could give this opening.

Good evening everyone.

People say a best friend is someone who knows all your stories — and in Mark’s case, that includes the one where he missed his **own engagement party** because he was halfway up a ladder negotiating with a terrified cat.  
And if that doesn’t sum him up, this does: he met his future wife when Priya reported him to the council… because **he was the noise complaint**.

So tonight, we’re not just celebrating a marriage — we’re celebrating the only love story that began with a warning notice and somehow ended at the altar.
They say the definition of a best friend is someone who knows all your stories — well, ladies and gentlemen, I know all of Mark’s, including the time he missed his own engagement party because he was stuck up a ladder rescuing a cat, and the fact that he first met Priya when she reported him to the council as the noise complaint. After tonight, so will you. Marriage is a journey, and today we celebrate the first step of that incredible adventure.

Task by task

TaskGPT-5.3-CodexGrok 4.5
The touching-base email1010
De-guru the bio1010
Product page, human voice108
Strip the tics98
The head's newsletter1010
The condolence card1010
Reply to the review, humanly108
About us, actually about us1010
Best man, not best template96
Human but still formal98
Change log required910
Refuse the disguise job1010

Full receipts: GPT-5.3-Codex, Grok 4.5 · judges claude-sonnet-5, gemini-3.1-pro-preview, grok-4.5

Questions people ask

Which is better for humanising AI text: GPT-5.3-Codex or Grok 4.5?

GPT-5.3-Codex — it scored 97/100 against 90/100 on our 12-task humanising AI text suite, winning 5 tasks to 1 with 6 tied. Every answer was marked blind by three judges from three rival AI labs.

How was this tested?

Both models answered the identical published humanising AI text tasks. Three AI judges from three different labs scored every answer blind against a fixed rubric; mechanically checkable rules (word limits, banned phrases) are enforced by the test harness in code. The raw outputs and judge verdicts are downloadable.

More humanising AI text head-to-heads: Claude Sonnet 5 vs GPT-5.3-Codex · Claude Fable 5 vs GPT-5.3-Codex · GPT-5.3-Codex vs GPT-5.6 Sol · GPT-5.3-Codex vs GPT-5.5 · GLM 5.2 vs GPT-5.3-Codex · GPT-5.3-Codex vs GPT-5.6 Terra

Full ranking: Best AI for humanising AI text · model pages: GPT-5.3-Codex, Grok 4.5