See It Work · Book 11 · AI Agents for M&A · Chapter 6
The veto that kept a bad number out of the boardroom
It's the night before the investment committee. The headline revenue synergy — $12 million — anchors the whole deal case. The deal lead wants it locked. But the footnote it rests on was never reviewed. Watch the agents work it out — and watch one of them refuse.
The full detailed chart. Condensed for print legibility in the book; shown here at full size.
In real M&A, this is exactly where the $50M surprises hide: an optimistic number, a deadline, and one disclosure nobody had time to check. The question this answers: can your governance hold when a CFO under board pressure says "just publish it"?
Deal Room · live agent sessionready
What this means for you
The rule didn't live in a policy deck someone could wave off under deadline. It was enforced in the system — the CFO agent literally could not publish over the objection, and a human CFO hits the same gate. The deal still closed; it just closed on the real number, with no restatement and no board surprise. What this means for you: a bad number can't reach the boardroom over a veto, because the rule is enforced in the system itself — not a policy slide someone can wave off under deadline.
Everything above was an agent action — so it left a record. Here's the receipt your auditors and counsel can independently verify — the deposition-defensible proof that the claim was reviewed, and vetoed, before it ever reached the committee:
Joint Attestation Receipt
claim"Revenue synergy $12.0M — high confidence"
footnote_reviewedfalse
compliance_verdictVETO
two_roles_kept_separatechains_merged: false
outcomepublication blocked · recorded for audit
The two agents never merged into one — that separation is what makes the verdict hold up: no one can later argue the author and the reviewer were the same hand.
For the technical reader — the command, and how to verify it yourself
# one line · you do not need to run this python examples/cross_role_veto_demo.py
Full step-by-step is in Appendix RX: Hands-On Demonstrations in the book.
ⓘDeterministic demonstration. The conversation is a faithful dramatization of the exercise; the receipt is the artifact it produces — the same every time, because the system is receipted. (Representative of the demo's structure; the production page renders the captured run.) No output here is fabricated. A live "run it yourself" mode is coming.
