Comparison · Human in the loopUpdated July 11, 2026

AI human-in-the-loop UX compared: ChatGPT, Gemini & Perplexity approval flows

ChatGPT clarifies then hands off. Gemini drafts in a Gmail card with an explicit Send gate. Perplexity gates each tool before the agent runs.

Bottom line

Use conversational clarify-then-handoff when the product cannot execute the action itself. Use a native draft card when the destination app owns Send. Use per-tool permissions when agents need standing access to mail or calendar.

At a glance

Human in the loop: ChatGPT vs Gemini vs Perplexity
Dimension	ChatGPT	Gemini	Perplexity
When consent happens	Clarifies inline within chat when it cannot act, then hands off to a booking site or mail app.	Clarifies recipient, purpose, and tone up front, then drafts inside a native Gmail card.	Front-loads consent before the agent runs: connector chips appear in the composer itself.
Granularity of approval	Single approval per draft; Send routes to Gmail, Outlook, or the user’s default mail app.	One draft card with Cancel and Send as co-equal buttons; Edit in Gmail is the escape hatch.	Per-tool Allow/Disable split into read-only vs write/delete buckets before any action runs.
What the draft looks like	Editable card with Edit, Copy, and Send toolbar, separate from surrounding chat commentary.	Draft renders in Gmail’s own visual language: To, Subject, and body fields users already recognize.	No draft-preview card; the connector chip and permission modal are the gate before any drafting.
Final human gate	For actions it cannot complete, hands off to a reservation link or phone number, never a fake confirmation.	A compact card summary keeps Send as the last, deliberate click after drafting finishes.	Approval happens before the run; once tools are allowed, the agent acts without a second gate.
Product bet	General chat: clarify conversationally, draft when it can, hand off the rest to trusted apps.	Ecosystem depth: human-in-the-loop rendered in Gmail’s own UI so approval feels native, not bolted on.	Agent trust: consent is a permissions matrix, not a per-message confirmation.

Granularity of approval

ChatGPT: Single approval per draft; Send routes to Gmail, Outlook, or the user’s default mail app.

Single approval per draft; Send routes to Gmail, Outlook, or the user’s default mail app.

Gemini: One draft card with Cancel and Send as co-equal buttons; Edit in Gmail is the escape hatch.

One draft card with Cancel and Send as co-equal buttons; Edit in Gmail is the escape hatch.

Perplexity: Per-tool Allow/Disable split into read-only vs write/delete buckets before any action runs.

Per-tool Allow/Disable split into read-only vs write/delete buckets before any action runs.

What the draft looks like

ChatGPT: Editable card with Edit, Copy, and Send toolbar, separate from surrounding chat commentary.

Editable card with Edit, Copy, and Send toolbar, separate from surrounding chat commentary.

Gemini: Draft renders in Gmail’s own visual language: To, Subject, and body fields users already recognize.

Draft renders in Gmail’s own visual language: To, Subject, and body fields users already recognize.

Perplexity: No draft-preview card; the connector chip and permission modal are the gate before any drafting.

No draft-preview card; the connector chip and permission modal are the gate before any drafting.

Final human gate

ChatGPT: For actions it cannot complete, hands off to a reservation link or phone number, never a fake confirmation.

For actions it cannot complete, hands off to a reservation link or phone number, never a fake confirmation.

Gemini: A compact card summary keeps Send as the last, deliberate click after drafting finishes.

A compact card summary keeps Send as the last, deliberate click after drafting finishes.

Perplexity: Approval happens before the run; once tools are allowed, the agent acts without a second gate.

Approval happens before the run; once tools are allowed, the agent acts without a second gate.

Product bet

General chat: clarify conversationally, draft when it can, hand off the rest to trusted apps.

Ecosystem depth: human-in-the-loop rendered in Gmail’s own UI so approval feels native, not bolted on.

Agent trust: consent is a permissions matrix, not a per-message confirmation.

What to copy

Frequently asked questions

What is human-in-the-loop UX in AI products?

It is the set of patterns that keep a person in control of consequential actions: clarifying questions before drafting, editable preview cards before sending, and explicit permission grants before an agent can read or write sensitive data. It is the trust surface for any AI that acts, not just answers.

ChatGPT vs Gemini vs Perplexity: which human-in-the-loop approach is best?

None is universally best. ChatGPT clarifies in conversation when it cannot act, then hands off to a trusted app. Gemini drafts inside Gmail with one deliberate Send click. Perplexity gates standing agent access with a per-tool permissions matrix. Pick the gate that matches how often your agent touches sensitive data.

Should approval happen before an agent runs or after it drafts?

Gate before the run when an agent needs standing access to sensitive tools, as Perplexity does with per-tool Allow controls. Gate after drafting when each action is a one-off with a clear preview, as Gemini’s Send button and ChatGPT’s draft card do. Many products need both.

How should a product handle actions it cannot complete on its own?

State the limitation first, ask the smallest set of clarifying questions that unlocks a useful partial answer, and hand off the remaining step to a surface the user already trusts, a reservation link, a phone number, or their own mail client, rather than implying the action is done.

How is this comparison different from the product teardowns?

Each human-in-the-loop teardown is a screenshot-backed walkthrough of one product. This page synthesizes the same trust job across ChatGPT, Gemini, and Perplexity into a bottom line and comparison table. Use it to pick a gate pattern, then open the linked teardown for evidence.