Meta is putting AI agents inside a billion WhatsApp conversations — and asking them to close the sale.

On Wednesday, Meta launched the Meta Business Agent globally, an AI agent that lives inside WhatsApp, Messenger, and Instagram and handles customer questions, recommendations, and orders on behalf of businesses. It's the global rollout of a product Meta has been quietly testing with small businesses in India since last month.

The key difference from earlier business chatbots: this one is built to take the money. Meta's head of product Naomi Gleit told Reuters the company wants the agent to "complete the payment, to process the booking, to place the order" — all inside the same chat window where the conversation started.

This lands in a market most American AI coverage overlooks. According to a 2025 Kantar study, 91% of online adults in India chat with a business on WhatsApp every week. Messaging isn't a support channel there — it's the storefront. Early results from Indian small businesses back that up:

• Overnight sales recovered: One plant-based brand said the agent handles inquiries that used to go unanswered, with conversion rates hitting 80–90%.
• No-code setup: A gifting business owner got the agent running in a few hours, and it now replies to customers in her voice.

Whether Western consumers will be as comfortable is a harder question. The 2026 Digital Trust Index from Thales found that only 23% of consumers trust companies using AI to handle their data — and trust in social media as an environment for any of this sits at just 9%. Meta is building transactional agents on one of the least-trusted categories of platform, for customers already nervous about both. And agents will find every weak seam in the systems they touch — Gleit herself acknowledged an early stumble where agent traffic exposed a bug that let people break into Instagram accounts, since fixed.

The business model is new for Meta, too. For large enterprises, the agent is priced on tokens consumed — the same model as OpenAI and Anthropic — layered on top of existing WhatsApp Business fees. Goldman Sachs projects token consumption could grow roughly 24-fold by 2030 as agents take over more transactions, which explains why Meta wants to own the meter.

While OpenAI and Anthropic spend the year fighting over who builds the smarter coding agent, Meta is walking into the conversations that already drive commerce for most of the world. The interesting move isn't the technology, it's the geography. Most American readers will hear about the Meta Business Agent the first time they get a slightly-too-helpful reply from a clothing brand on Instagram. A tea-stall owner in Pune will have been using it for a year by then. If agents really do become the new storefront, the company with the biggest one already built is the one nobody in Silicon Valley spent the year talking about.

Last week, Trump signed a scaled-back AI executive order that leans heavily on voluntary cooperation between the government and the big labs. This week, OpenAI published a policy paper arguing the feds should go further — requiring the most powerful AI models to be evaluated by a federal agency before they're released to the public.

The agency is CAISI, the Center for AI Standards and Innovation, which sits inside NIST and tests frontier models for safety and security risks. OpenAI already shares data with CAISI voluntarily, but now it wants that arrangement made mandatory for everyone building at the frontier. OpenAI's head of global policy, Chris Lehane, told Politico the Trump order validated the safety work OpenAI has already been doing, and that CAISI is capable of running serious evaluations. The policy paper goes a step further: policymakers should require frontier models to clear a CAISI evaluation before public release, with the agency recommending mitigations rather than approving or denying launches.

This is the part most readers wouldn't expect. OpenAI is the company everyone assumes wants Washington to stay out of its way. Instead, it's openly asking for a tougher regime than the one the White House just delivered.

The skeptical read is that OpenAI is asking for regulation it already complies with — which would mostly inconvenience everyone else. Sanchit Vir Gogia, chief analyst at Greyhound Research, put it bluntly: "Governance of this shape can become a moat dressed as maturity." There's also a practical problem: the Institute for Progress estimates CAISI would need at least $84 million a year to carry out this kind of mandate. Its current budget is closer to $20 million. The agency OpenAI wants as its overseer can't actually do the job yet.

OpenAI isn't alone in pushing for guardrails. Last month, Altman and Anthropic's Dario Amodei signed a joint letter calling for stronger rules on AI-assisted biosecurity risks — one of the rare moments the two CEOs have agreed on anything publicly. The frontier labs are starting to sound remarkably similar when it comes to what they want from Washington.

OpenAI asking for tougher rules sounds like a company breaking ranks with its own interests, until you remember that the company already passes the test it's proposing. A rule that says "every frontier lab must be evaluated by CAISI" is a rule OpenAI complies with on Tuesday and a rule a smaller competitor spends a year preparing for. The cleaner version of this story isn't that OpenAI suddenly wants to be regulated. It's that OpenAI figured out which regulation it can live with, and decided it would rather help write that one than wait to see what California, New York, and twenty other states write instead.

For months, workers have been saying it out loud: companies are blaming AI for cuts they were going to make anyway. The data finally caught up.

The latest Challenger, Gray & Christmas report tells the story clearly: AI-linked cuts jumped from 8.7% of all announced layoffs in February to the top cited reason by April. May brought 82,744 total cuts — the highest May figure since 2023.

• Tech is leading the charge. Through April, the sector announced 85,411 layoffs, up 33% year over year, accounting for nearly 39% of all April cuts on its own.
• The script is predictable. Amazon CEO Andy Jassy told employees the company would rightsize its corporate footprint "as we get efficiency gains from using AI extensively" — shortly before cutting roughly 14,000 corporate roles.
• The savings haven't materialized. A PitchBook analysis found that 40% of executives expected AI to deliver 15–20% IT cost savings. Most achieved only 5–15% in specific functions.

The question is whether any of this is actually about AI — and analysts increasingly say no. Forrester's J.P. Gownder calls the pattern "AI washing": executives cite AI because it makes them sound innovative, when the real driver is usually a softer revenue forecast or investor pressure. MIT Sloan's Paul Osterman put it more bluntly — AI is a "convenient justification for pre-planned workforce reductions," a way to depersonalize difficult decisions.

Sam Altman has argued the opposite, telling CNBC that organizations deeply integrating AI tend to show net job growth — though he cautioned the correlation shouldn't be mistaken for causation. Convenient framing either way.

Companies are cutting first and hoping the productivity shows up later. Even Andy Challenger, whose firm publishes the data everyone's quoting, has noted there is "no evidence of an AI-driven jobpocalypse" and that productivity gains remain speculative.

Every restructuring era gets its own villain. In the '90s it was globalization. In the 2010s it was automation. In 2026 it's AI, and it's the most useful one yet because nobody can really prove or disprove the link. If you cut 10,000 people and say AI did it, you sound forward-looking. If you cut 10,000 people and say demand softened, you sound like you missed your number. Most of the CEOs reaching for AI as the reason haven't measured the productivity gains. They're reaching for it because it's the only excuse that comes with a stock bump attached.