There's 'No Evidence That Having Sex With Robots Is Healthy' and No Evidence That It's Not

There's just no evidence about sex robots period, because at present they don't really exist. But that hasn't stopped folks from freaking out...


Eduardo Contreras/TNS/Newscom

There's "no evidence that having sex with robots is healthy," The Washington Post wants you to know. Similar headlines grace the pages of USA Today ("There's no evidence having sex with robots is healthy, report says"), The Verge ("no evidence that sex robots provide health benefits"), CNBC ("Sex robots offer little evidence of any health benefits"), and many other outlets, sometimes with an added dose of alarmism ("Sex robots could empower pedophiles and sex offenders") or millennialism ("There's literally no research proving sex robots are good for society").

These headlines are all true, more or less, but they omit an important fact: There's also "literally no research" showing that sex robots unhealthy, that they don't hold therapeutic promise, or that they will be bad for society. There's just no evidence about sex robots period, because at present they don't really exist.

The authors of a new article in BMJ Sexual & Reproductive Health—the study behind all those headlines—admit as much when it comes to the lack of evidence. "We found no reports of primary data relating to health aspects of the use of sex robots," write Chantal Cox-George, a doctor at London's St. George's Hospital, and Kings College Professor of Women's Health Susan Bewley.

They are trickier when it comes to the current state of sexbots. The definition of sex robot that they offer doesn't include artificial intelligence, or even robotics, as necessary components: "Sex robots are realistic mannequins with variable ages, appearances and textures, and customisable oral, vaginal, and anal openings." But basically everything else in the paper implies that we are talking about not just mannequins but moving, talking, artificially intelligent creatures.

Cox-George and Bewley assert that there are currently four companies selling "adult sexbots," but they don't explain anything more about the products these companies make. If they had, readers would learn that the closest thing to sexbots at present are simply life-size dolls with very basic digital flourishes. (Think Siri inside a RealDoll, not the sentient android denizens of Westworld.)

But maybe it makes little difference: A lack of actual sexbots in the world hasn't stopped decades of folks freaking out over the possibility of randy AI-enabled companions. The only difference these days is a persistent (and wrong) belief that they are already here or right around the corner.

The Post article quotes all sorts of sexbot alarmists (including the founder of a Campaign Against Sex Robots), as well as pop-media hype about the therapeutic potential of sexual and romantic relationships with robots. Cox-George and Bewley's article affects an air of more serious scholarship, but it cites only the same sexbot speculation you can find shared on your Facebook feed or gracing Google News every few months.

Using Google and PubMed searches, Cox-George and Bewley relied on results from HuffPost, The Guardian, The Atlantic, the Daily Mail, The Independent, The New York Times, and other consumer-facing media, along with information from the Campaign Against Sex Robots and Foundation for Responsible Robotics websites. From these sources, the doctors absorbed the same old (highly unscientific and culture-war-tinged) controversies surrounding not just sex robots but also sex workers, prostitution, pornography, virtual reality, and sexual assault, then summarized these controversies briefly.

The article does not unveil any new scientific evidence about sex robots, does not synthesize previous evidence, does not even offer any compelling new speculation on the topic. The authors simply regurgitate pop-media coverage and controversies regarding sex dolls and sexbots and then suggest that more research is needed.

Until "robust, scientific, and ethically acceptable research trials" can be conducted, medical professionals should "reject the clinical use of sexbots," they conclude. They also suggest that "doctors might be advised to avoid using sexbots themselves, given police interest, prosecutions, and the potential negative impact on public trust."

But whatever doctors do with sexbots—on their own time or in clinical practice—public interest in them isn't going anywhere. For a more positive view of their potential, check out this feature from Reason's 2015 "Hi, Robot!" issue.