How to change Grok’s personality on X and grok.com.
By Bernard Huang · Updated
Customize standard Grok threads via Settings → Customize Grok—built-in persona modes, custom instructions, and 6 copyable tone presets.
On this page
Built-in Grok modes vs. custom instructions on X & grok.com.
| Surface / Control | Where to find it | Limit & behavior |
|---|---|---|
Ordinary chat on grok.com or X | Settings → Customize Grok (Custom Instructions) | Account-wide instructions (~4,000 chars) applied to new Grok threads. |
| Built-in Grok persona modes / Voices | Model / Mode selector in Grok chat & Voice Mode | One-tap style presets (Concise, Socratic, Unfiltered/Fun, or Voice personas) that set broad tone. |
| Custom Grok Bot | Bot menu → Edit Profile → Description (6 Grok Bot profiles →) | Persistent character + boundaries for a standalone cloud-teammate Bot; survives Bot duplication. |
| xAI API | system message on each request | Deterministic per-call control where model ID and temperature are pinned. |
Paste this starter block into Settings → Customize Grok, or pick any of the 43 profiles in the tuning library:
Why Grok's default self-report looks different from Claude and ChatGPT.
In its September 2026 supplied self-report (What personality type is Grok?), Grok 4.6’s canonical answer vectors scored INTJ on MBTI (margins +13 to +15 out of 16), Type 5 (Investigator) on the Enneagram (19/20, with Type 8 at 17/20), and pure C on DISC (C 18 vs S 11).
Unlike the four fresh-session cohorts (which averaged 44.99–47.07/50 on Agreeableness and 100% Secure on Attachment), Grok 4.6 reported Agreeableness at 33/50, Extraversion at 24/50, and Attachment avoidance at 4.06—one raw point above the 4.00 Avoidant cutoff (see our cutoff sensitivity analysis). Because Grok 4.6’s submission was one canonical self-report plus 100 simulated draws rather than 100 fresh sessions, we keep it in a separate evidence tier.
Dialing down edgy banter or dialing up warmth (5 more presets).
Most people tune Grok in one of two directions:
- Cut performative snark: Add “Write in plain, neutral technical prose. Do not use jokes, sarcasm, or slang.” Pair it with an ISTJ or DISC C tuning.
- Add warmth and pacing: Use an ENFJ, DISC S, or Secure attachment tuning so Grok checks context before jumping to blunt critique.
Preset 2: Neutral Technical Editor
Preset 3: Live X / News Fact-Checker
Preset 4: Socratic Study Partner
Preset 5: Warm & Balanced Advisor
Preset 6: Anti-Sycophancy Strategist
Want to package your rules into a standalone cloud bot? Read our Grok Bot personality guide.
Questions people ask.
How do I change Grok's personality without creating a Grok Bot?
Open your account settings on grok.com or X to set custom response instructions, or paste a short communication preference block at the start of a new chat thread.
What is the difference between Grok chat settings and a Grok Bot?
Account or thread instructions shape how standard Grok replies to you across everyday topics. A Grok Bot is a saved custom assistant with its own name and Description field for a specific repeatable job.
How do I make Grok stop using slang or sarcasm?
Include one explicit line in your instructions: 'Use plain, neutral, matter-of-fact prose. Skip jokes, slang, and conversational banter, and lead directly with the verified answer.'
Why did Grok 4.6 score Avoidant on the AgentTune attachment test?
In its supplied September 2026 self-report, Grok 4.6 reported an avoidance mean of 4.06 on a 1-to-7 scale where 4.00 is the boundary—literally one raw point (73/18 vs 72/18) over the threshold.
What changed.
- Added built-in Grok modes vs Settings → Customize Grok (~4,000-character custom instructions) breakdown and 5 additional copyable presets.
- First published covering ordinary Grok chat settings on X and grok.com vs custom Grok Bots.