Meihaku

AI Support Readiness · Pre-flight your agent

You can’t launch AI support on knowledge you haven’t audited.

Stop spending weeks manually reviewing docs, tickets, macros, and internal notes. Meihaku maps your existing support knowledge against real customer intents, surfaces gaps and conflicts, and gives your team an approved answer set before an AI agent talks to customers.

Launch readiness · click to watch the overview

Product overview

Show the launch blockers before the agent talks to customers.

The audit loop is deliberately narrow: connect the knowledge you already have, map it to customer intents, resolve blockers, then approve only what your team can defend.

01

Connect

Docs, tickets, macros

02

Map

Customer intents by readiness

03

Resolve

Gaps and conflicts

04

Approve

Approved answer set

Step 01

Connect read-only support knowledge

Point Meihaku at the docs, tickets, macros, and notes your team already relies on.

Meihaku sources screen showing connected Google Drive and Zendesk source bundles

Step 02

Map customer intents

See which customer questions are covered, conflicted, or missing before your AI support agent handles them.

Meihaku topics screen showing billing, plans, and account-management intents with gap, answered, and conflict labels

Step 03

Resolve source conflicts

When two sources disagree, reviewers see both claims side by side and choose the canonical answer.

Meihaku conflict screen comparing two refund-window sources

Step 04

Approve cited answers

Only source-backed answers become part of the approved answer set your downstream agent can use.

Meihaku answer screen showing an approved service outage answer with supporting evidence

Why readiness comes first

Source quality decides AI quality.

If the source knowledge is missing, stale, or contradictory, your AI agent can still answer with confidence. Meihaku addresses the source-readiness layer; runtime moderation and agent-platform guardrails handle the rest.

  • Air Canada

    Tribunal ordered the airline to honor a bereavement-fare policy its chatbot invented. C$812 + ruling that the chatbot is part of the website.

    Moffatt v. Air Canada, BCCRT 2024
  • Cursor

    Support bot “Sam” invented a one-device-per-subscription policy that did not exist. Subscription cancellations followed.

    Fortune, April 2025
  • DPD

    Chatbot swore at a customer and wrote a poem calling DPD “the worst delivery firm in the world.” 1.3M views on X.

    TIME, January 2024

“Air Canada is responsible for all of the information on its website, regardless of whether said information comes from a ‘static’ webpage or a chatbot.”

BC Civil Resolution Tribunal, Moffatt v. Air Canada, 2024 BCCRT 149. Read the decision

Readiness, not chat

Audit the knowledge. Approve the answers. Ship a defensible boundary.

Most teams discover the real work after the chatbot project starts: outdated docs, conflicting policies, missing answers, and tribal knowledge trapped in tickets. Meihaku turns that cleanup into a structured launch-readiness map.

Point Meihaku at your support knowledge

Connect read-only docs, tickets, macros, and notes. Meihaku keeps citations tied to source evidence.

Meihaku shows where launch will stall

It matches customer questions to source evidence, drafts cited answers, and flags the gaps and conflicts blocking safe automation.

You approve what ships

Review each draft with sources attached. Approved intents become the answer set your AI agent is cleared to handle. Unsupported intents stay blocked.

Built for the support manager carrying launch risk

Launch with approved knowledge, not hopeful retrieval.

Coverage you can defend

See which intents have approved answers, which are blocked, and which need a human before launch. The boundary is explicit.

Every answer cites a source

Review the exact policy line, macro, or article behind each draft before approving it.

No blank-page rewrite

Meihaku does the reading and first pass from the materials you already have. Your team only greenlights what is safe to ship.

Blockers surface before customers see them

Missing or conflicting intents stay out of automation until your team has source evidence or a handoff path.

Frequently asked

Questions support and compliance teams ask before they buy.

How do I stop my AI support agent from hallucinating?

Hallucinations usually happen when the model can’t find a clear answer in the source material it retrieves from. Meihaku audits your support docs, tickets, macros, and notes for the gaps and conflicts that force the model to guess, then gates which intents your agent is cleared to answer. Many teams pair Meihaku as the pre-flight check with runtime tools like Cleanlab or Ada-style monitoring as the in-flight safety net.

I haven’t picked an AI support agent yet — is Meihaku still useful?

Yes — and it’s the cleanest case to start with. Meihaku audits your support knowledge before you commit to a vendor. You’ll know in advance which intents are well-supported, which need source cleanup, and which can’t be answered safely by any agent. Most teams find the audit changes which agent they pick, because some handle gap-prone intents better than others.

How is Meihaku different from AI support agents like Intercom Fin or Decagon?

Meihaku is the readiness layer that runs before and alongside AI support agents, not a replacement. Tools like Intercom Fin, Decagon, Sierra, and Maven generate answers at runtime. Meihaku audits whether your knowledge base actually supports those answers in the first place, surfacing conflicts, gaps, and ungrounded intents before they reach a customer. Most teams use Meihaku to pre-flight an AI agent rollout, then keep it running as a governance layer.

What sources can Meihaku read today?

Google Drive folders containing your support policies, SOPs, macros, and FAQ docs, plus Zendesk source bundles. Intercom, Notion, Confluence, and broader helpdesk connectors are on the near-term roadmap. Meihaku connects read-only and never writes back to the source — what you see in the audit is exactly what your team approves.

Does Meihaku replace my AI support agent?

No. Meihaku is complementary. Your AI agent (Fin, Decagon, Sierra, custom) still answers customers. Meihaku decides which intents are cleared for that agent to handle, with cited evidence, and exports the approved answer set. Unsupported or conflicting intents stay out of automation until your team has source evidence or a handoff path.

Can my team approve answers before customers see them?

Yes — that’s the core workflow. Meihaku drafts cited answers from your sources, your team reviews each one with the source line attached, and only approved intents become the boundary your AI agent operates inside. Nothing reaches production until you greenlight it.

What if my knowledge base is a mess?

That’s the most common starting point — and exactly what Meihaku is designed for. You don’t need to rewrite anything before connecting. Meihaku reads what you have, drafts cited answers from the strongest evidence, flags gaps and conflicts, and lets your team triage them in priority order. The first audit usually surfaces the small set of intents blocking most of the customer-risk conversations.

Is my support data used to train AI models?

No. Customer data is processed in your isolated workspace. We don’t train foundation models on your content, and we don’t share data across customers. Source content is processed transiently and not retained beyond what’s needed for citations.

Get your AI support readiness map.

Connect support knowledge read-only. Meihaku finds the gaps and conflicts blocking launch, then helps your team approve what the AI is cleared to say.

Get started