Music streaming service Spotify is taking its advertising tactics up a notch, with a new feature that will place ads based on the tone of your playlist.

The product, called Playlist Targeting, leverages Spotify’s listener data to push ads. In a press release, the company wrote:

Brands can now target audience segments based on who they are (age & gender, geography, language), what they’re listening to (playlist, genre), and when and how they’re listening (time of day and by platform/device).

While super-targeting is very hot among advertisers, placing ads based on someone’s mental state seems a little reminiscent of Facebook’s heavily criticized mood experiment.

For those who don’t remember, Facebook injected the feeds of 700,000 unwitting users with negative content to see if they would react with negative posts of their own. At the time researchers noted, “When positive expressions were reduced, people produced fewer positive posts and more negative posts; when negative expressions were reduced, the opposite pattern occurred.” So mood manipulation seemed very much achievable.

People were especially angry because the study was conducted without the consent of the users involved.

Though Spotify’s new mood-targeting feature is not a study and isn’t subject to the same ethical standards, it does deserve closer examination.

The link between consumer mood and buying behavior has long been studied. But the effects of newer forms of adtech, like serving ads based on mood, have not. We don’t yet have standards to determine whether this kind of practice is ethical. For instance, we know it is unethical and illegal for businesses to misrepresent statistics to tell a story about their company that isn’t true. But it took time for those laws and standards to develop, and with technology moving more rapidly than ever, it’s hard not to wonder if ad targeting at the mood level doesn’t have the potential to do damage. By allowing mood-targeting ads, are we opening up the door to potential unethical emotional manipulation in the interest of selling products?

Imagine you’re post-breakup and you’re listening to “Ultimate Breakup Playlist.” All of a sudden Spotify serves you an ad for Celesta, an antidepressant. Wouldn’t that feel just a little bit invasive and perhaps taking advantage of your weakened state? Or would you feel like maybe it is time to go on antidepressants and this advertisement was helpful?

The company says its Audience Targeting platform leverages data about your gender, location, behavior, mood, music taste, trends, and, most off-putting of all, “need states.”

There aren’t any clear answers yet on whether mood targeting has the potential for harm or if it’s just a means to better serve consumers the products they want. But, as companies grow their ability to track down users and identify them by everything but their name, we need to stay aware of the potential for abuse.