Even while some media organizations roll out new online subscription plans, the Internet continues its steady drift toward a business model built overwhelmingly on money not from readers, but from advertisers. It’s advertising that’s emerging as the revenue source that everybody, from Facebook and Google to newspaper websites and gadfly bloggers, wants a piece of.

That raises perennial questions of media economics and ethics: What limits should publishers put on advertiser influence? How far should they go in shaping their content to enhance its value as a delivery van for paid persuasion?

And how plainly should they cordon off the messages they’re paid to carry to distinguish them from content that’s shaped independently by their own staffs, whether reporters, aggregators or curators?

The latest mini-scandal involving advertising over-reach doesn’t come from an online startup. It’s from a venerable and highly respected publication, the Atlantic, founded in 1857 by Longfellow and Emerson.

In mid-January, the Atlantic posted a lengthy, self-congratulatory epistle from the organization called the Church of Scientology, which is variously assailed as a tax dodge and a loopy cult, praised as a source of spiritual rebirth, and chronicled as a refuge for disaffected Hollywood luminaries.

Now, running ads from controversial sponsors has long been routine among commercial media. But here, controversy erupted over the way the Scientology content — which was soon taken down — drew online comment that was suspiciously glowing, and over the way the item looked.

Although it ran under a small sig, “Sponsor Content,” the article otherwise seemed — in its typeface, composition, placement of photos, and overall look and feel — just like any other Atlantic article.

That had to be intentional, which takes us to the point of this column. The ad was an example of what’s now called “native advertising.” Native advertising refers to paid messaging that is created to resemble the articles and assorted editorial stuff in which it’s embedded.

A native ad is deliberately crafted to blend in with, and with luck, to be indistinguishable from the surrounding content. In that respect, it isn’t really “native,” it’s camouflaged.

Now such crypto-editorial ads, by tradition, carry some identifying tags to warn the reader they’re there only because somebody bought the space. But labels don’t really matter. There’s still an irreducible element of subterfuge to the whole enterprise.

For starters, such labels are used sparingly, and typically consist of some more or less opaque euphemism: “sponsored content” is a good example; “advertorial” (which few civilians recognize as English) is often used. Rarely do you see the straight-up, unmistakable alert, “Paid Advertising.” That would be clear. That would be a turn-off.

But more important, the whole purpose of the undertaking is squirrelly: It’s to appropriate the format of the surrounding publication and harness its credibility to strengthen the authority and persuasiveness of the advertising.

“Native advertising,” in short, is all about deception. You, as the reader, are encouraged to perceive the messages as something other than what they are. And even if, at some level, you understand they weren’t put together by the magazine’s staff, you’re still expected to see them as partaking of the magazine’s trustworthiness, and as deserving something of the same regard.

In the aftermath of this embarrassment, some of the criticism the Atlantic received was surprising, and focused on how poorly the magazine played the “native” game: Nobody, it was said, could mistake the Scientology missive for an article the Atlantic might actually have run. The magazine erred, Charlie Warzel of AdWeek wrote, by “licensing uninteresting and bizarre content that falls well outside the walls of the magazine’s brand.”

Economist magazine managing director Paul Rossi advised publishers to start thinking like ad agencies and get better at concealment: “The real issue is how do you make content that’s compelling to a reader that doesn’t feel like an ad. That’s the real challenge.”

So the problem was that the deception was inept and easily detected. If that’s it, the solution, it seems, would be to integrate paid persuasion seamlessly into editorial offerings, so that reader resistance is allayed and the ads go down smoothly.

But that’s not it. The more basic problem, as Columbia Journalism Review’s Dean Starkman wrote, is an “existential” one. In that regard, “native advertising” succeeds precisely because it’s so skillfully wrought that it’s imperceptible as advertising, and joins the flow of information and commentary that readers accept as part of the chronicle of topical realities that they trust the news media to provide.

Trouble is, at that point, it’s the entirety of the content that has truly gone native. And corruption has taken a mighty step forward.