On June 3, Meta launched something worth paying attention to. It is called Meta Business Agent. It is an AI agent that any business can deploy on WhatsApp or Messenger in minutes. The agent handles customer questions, collects details, books appointments, and escalates to a human when the conversation needs one. Meta’s own framing: “as if they had an infinite team behind them.”

That framing is worth sitting with.

This is free infrastructure, available right now, to every business with a WhatsApp or Messenger presence. And WhatsApp alone has more than 200 million business accounts worldwide, according to Meta’s most recent earnings disclosure.

Distribution Changes the Baseline

The technology here is not new. AI agents that answer customer messages have existed for years. What changed is the distribution layer.

When a nail salon in Jakarta or a landscaping company in Phoenix can have a 24/7 AI agent responding to every customer inquiry within seconds, the bar for “showing up” just moved. Permanently. The AI can respond while the person is away.

That is the part most people will miss. This is about a platform with 3.3 billion daily active users putting an AI agent one button away from every business on the platform. The distribution makes this structural.

The gap just got visible at every scale

For the past two years, the conversation about AI adoption has mostly lived at the enterprise level. Which companies are deploying agents internally. Which ones are building agentic workflows. Which ones are falling behind.

Meta Business Agent moves that conversation down to the level where most business actually happens. Small and mid-size operations. Service businesses. Local companies. The ones that answer phones, respond to inquiries, and book appointments. The ones where a missed message at 9 PM means a lost customer to whoever replies first.

Here is the pattern that will play out over the next six months. A portion of businesses will turn this on within weeks. An AI starts handling first-touch customer conversations around the clock. Leads that used to fall through the cracks get caught. Response times drop to seconds, at any hour.

The rest will hear about it, file it under “something to look into,” and keep doing what they have always done.

Six months from now, the businesses that activated early will have data. They will know which questions come in most often, what times customers reach out, which conversations convert. That information compounds. It feeds better responses and sharper offers. The businesses that waited will still be starting from zero.

I keep calling this Belief 2 — the gap between who acts and who waits. It has never been cheaper to close. Press the button or don’t. That is the whole decision now.

Why this matters beyond small business

If you run a mid-market or enterprise organization, you might think this does not apply to you. You have a customer success team. You have a support platform. You have processes.

But your customers are about to have a new expectation baseline. When every small business they interact with responds instantly through AI, they will wonder why your company takes 24 hours to acknowledge a support ticket. Every business your customer interacts with helps set the bar.

Cognizant and Snowflake announced at the same summit this week that their Cortex-powered AI agents have already handled 1.3 million requests across 2,250 users and reclaimed 1,300 hours of manual work. That is one deployment inside one company. Meta just opened the same category of capability to 200 million businesses simultaneously.

The scale of this distribution is the point. The technology question (“Can AI handle customer conversations?”) was answered two years ago. The distribution question (“Can every business access it?”) just got answered this week.

What to do with this

If your business has a WhatsApp or Messenger presence, the move is simple. Turn on Meta Business Agent. Today. The setup takes minutes and the downside is functionally zero. You can always adjust or turn it off.

If your business operates at a larger scale, the signal here is the expectation shift happening underneath you. Your customers are being trained by every interaction they have. When instant, intelligent responses become normal everywhere else, your response time becomes your weakness.

The businesses that treat this as a “nice to have” will look up in twelve months and find that instant AI response is table stakes. The window where this felt optional is closing faster than most people realize.

Meta just mass-produced the on-ramp to AI agents. And most businesses will still drive past it.