Skip to main content

Step 1 of 3

Choose your path to get started

30s - 2 min
1

Choose Path

What's Next

Standard path: immediate setup (30s) • Custom path: answer ten or less questions (2 min)

Back to providers
Grok logo

Humanize Grok

Choose your path to get started

How would you like to humanize your AI?

Choose the option that works best for you

providerDetailSeo.headingTemplate

Grok is the model with a personality. It will take a position, crack a joke, and skip the 'as a large language model' apology. That makes humanizing it less about removing AI tics and more about pointing the existing voice at the work you want.

providerDetailSeo.defaultsHeading

Looser. Opinionated. Casual register by default. More willing to be funny — or wrong — than the others. Closer to how a person texts than how a press release reads.

providerDetailSeo.bestForHeading

  • Opinion pieces, social posts, anything that benefits from a take
  • Casual writing — group chats, Slack messages, friendly emails
  • Drafts that need life — Grok overshoots, you trim, the result feels human
  • Real-time topics — Grok has live X data access where relevant

providerDetailSeo.pitfallsHeading

  • Too casual for formal contexts — the humanize prompt should set the register explicitly.
  • Will take a position you didn't ask for — instruct it to flag opinions vs. facts.
  • Snark can leak into unintended places — 'be warm, not snarky' is a useful line.

providerDetailSeo.promptTipsHeading

  • Grok's custom instructions field is in Settings. Same idea as ChatGPT.
  • If you want it more measured, set a tone explicitly ('professional, warm, no jokes').
  • Grok responds to vivid examples better than abstract rules ('Write like a thoughtful friend explaining over coffee').
  • The Fun mode vs. Regular mode toggle changes baseline behavior — the humanize prompt should account for which mode you use most.

providerDetailSeo.faqHeading

Does this require X Premium?
Grok 2 and above require an X Premium subscription. The humanize prompts work across all available Grok models.
Will it still be funny?
If you let it. The prompt can dial humor down without removing it — 'occasional wit, no forced jokes' works well.
What about factual accuracy?
Grok hallucinates at roughly the rate of its peers. The humanize prompt doesn't change that — it shapes voice, not truth.

Help us improve this page

Found an error or have a suggestion? We'd love to hear from you.