Not HR advice. Not legal advice. This page is not a substitute for counsel, an HR partner, your handbook, employment counsel in your jurisdiction, or your judgment. It describes a product ritual for hole-finding on a people message you already drafted. Nothing here creates an attorney-client or HR-advisory relationship.
Late day. A founder or manager has the calendar invite blocked, the termination note or hard exit message open in Docs, and a chatbot thread that already made the language feel cleaner: shorter, calmer, somehow inevitable. The person has not heard it. That is the moment this page is for.
People searching "before you fire" or "before you let someone go" usually get progressive-discipline checklists, PIP templates, and "document everything" advice. Those matter. This page is narrower. It treats the people decision and the message that carries it as a social commit you are about to speak into someone else's life, and it is about pressure-testing the AI-shaped read of that packet before the meeting starts.
This is not before you hire (offer letter going out). It is not before you send the letter (demand or hard notice to the other side). It is the inward people move: ending employment, or delivering the message that makes that ending real.
pingpong is a sequential review product for that ritual. Default chain: Grok, then Perplexity, then ChatGPT, then Gemini, then Claude. The first model drafts. Each later model sees your question, the prior answers, and a review frame that can agree, correct, restructure, or reject a weak premise. Brand: pingpong at pingpongit.com. Not getpingpong.ai.
Definition: What is pingpong. Mechanism: How it works. Role pages: For executives, For founders. First run: How to run a pingpong.
Your first eligible web review is free. When you need more, web Plus is $19.99/month and web Pro is $124.99/month. On iOS the listing shows three free pingpongs, then Plus at $24.99/month or Pro at $59.99/month. Start the free web review on pingpongit.com, then open Plans when you are ready for Plus or Pro. Full table: pricing.
Why a people message is a different kind of commit
A private note to yourself is cheap to rewrite. A message spoken in a room, or sent as the formal note after that room, is not. Once the person hears it, the words become the story they tell themselves, their family, a future employer, and sometimes counsel. Softening the reason later looks like you were never sure. Overclaiming a performance narrative trains them to distrust every claim that follows. Silence on logistics (access, pay timing, references) reads as contempt even when you meant care.
Managers often ask one model to "make this termination more professional" or "help me say this kindly." The model returns calmer prose that preserves the original bet to proceed. Socially pleasant. Thin when the next action is the meeting. Related habit for irreversible moves in general: Catch AI mistakes before you commit.
What usually hides in the people packet
The dangerous parts are rarely typos. They are soft premises dressed as certainty. A performance story that does not match what you wrote in reviews. A timeline that skips a coaching step you told yourself you finished. A comparison to peers that collapses under one hard question. A tone that reads as kindness to you and as ambush to them. A separation reason that is sharper than what your counsel or HR partner would clear. A "facts" paragraph that quietly invents a meeting nobody can prove. A promise about references or future help you will not keep under pressure.
One chatbot grading its own exit draft rarely catches that pattern. It softens edges and keeps the structure. Later labs, reading a concrete packet without being told to flatter your resolve, are a different social object. Framing: AI second opinion. Sycophancy angle: Debias AI and sycophancy research.
How the five-model handoff works on a people decision
Paste one clear question, not a vibe. Example shape: "We plan to let ROLE go on DATE for REASON as documented in ATTACHED notes, with MESSAGE as drafted, delivered by METHOD, with LOGISTICS as listed. Where is this weak, overclaimed, inconsistent with the written notes, or expensive to unwind after they hear it?" Attach the draft message, the chronology you trust, and any counsel or HR notes when you can. Do not paste secrets you should not put in a third-party tool.
Grok goes first. Perplexity reviews with that draft in view. ChatGPT, Gemini, and Claude follow in order. Each pass can keep, fix, or refuse. You get a final answer and can open earlier passes. The product bet is not that five models invent the right HR policy. It is that skipped edge cases and soft premises are harder to ship unnoticed when later labs have to look at them.
Architecture: Sequential vs parallel AI. Category map: Multi-model AI review. Fair single-model contrast: pingpong vs ChatGPT.
What to listen for in the middle passes
Treat independent convergence as a stronger signal than polite agreement. If three later models keep challenging the same clause (reason that does not match the written record, missing logistics, tone that reads as ambush, claim about peers you cannot defend, timeline that skips a step you said you took), that clause is your checklist, not a reason to force a synthetic consensus.
Treat unresolved split the same way. One model may want a quieter coaching path first; another may want a narrower reason before any meeting. You still decide. pingpong does not replace your judgment, your HR partner, or counsel. It pressure-tests the AI-shaped people story you were about to trust.
Decision reliability framing (not uptime SLAs): Extreme reliability. Clarification of the "insurance" metaphor: Decision insurance.
When to skip the chain
Skip it for exploratory notes to yourself, early drafts nobody will see, and role-play scripts with no real person attached. One strong model is enough when being wrong costs almost nothing. Save the free eligible review for the message that is hours from a real conversation, or the formal note that will follow it. Role pages if the decision is not yours alone: executives, founders, teams, lawyers.
Related commits
Cluster hub: before you commit. Same handoff, other irreversible moves: before you hire, before you change pricing, before you send the letter (not legal advice), before you reply to the board, before you sign the contract (not legal advice), before you launch, before you post, before you partner (not legal advice). Pattern page: catch AI mistakes before you commit. Chooser: when to use pingpong. Continue on Plus or Pro.
Plans
Your first eligible web review is free. Further web reviews need a subscription. Web Plus is $19.99/month. Web Pro is $124.99/month. iOS lists three free pingpongs, Plus at $24.99/month, and Pro at $59.99/month (yearly options appear on the App Store). Confirm the live plan at checkout. Details: plans and pricing. Related: Is pingpong worth it.
Run it before the message goes out
Take the people decision, the draft message, and the logistics you almost locked. Restate them as one decision question. Run the default pingpong order once. Keep what survives. Fix what later models break. Escalate what they cannot settle to a human who owns the relationship and to counsel or HR when the stakes require it. Start with the free eligible web review on pingpongit.com. If the chain earns a place before your next irreversible people message, choose Plus or Pro under Plans (web Plus $19.99, web Pro $124.99; iOS Plus $24.99, Pro $59.99). Or start on the App Store.
Again: not HR advice. Not legal advice. Not medical or financial advice either. Hole-finding on your draft is not a substitute for counsel or an HR partner.