RN
Start with the correction, because the question in the title contains a false assumption.
Women do not, as a group, prefer gay porn. On the platform that generates most of the data anyone cites about this, the category women view most is lesbian, with Japanese, MILF and anal filling out the top of the list. If you were looking for evidence of a mass female preference for men having sex with men, it is not there.
What is there is stranger and more worth explaining. Across Pornhub's recent Gay Pride Insights reports, the share of the gay category's audience recorded as women has sat somewhere in the neighbourhood of 43 to 50 percent, trending upward year on year. Trade coverage of those reports does not agree on which figure belongs to which year, so treat the range as the claim rather than any single number.
Roughly half. Of a genre named after, and marketed to, gay men.
What that number is, and what it is not
Before building anything on it: this is one company's internal analytics, gender is self-selected by the visitor, and nobody verifies it. It describes people who used a particular free tube site while logged in or profiled, not women, and certainly not women who pay for anything.
Nor does it tell you what the viewing was for. A category tag counts a page view. It cannot distinguish arousal from curiosity from an accidental click from a couple watching together from someone doing exactly what the headline assumes. Any explanation offered below is an explanation of a pattern in traffic logs, which is a much weaker object than it sounds.
With that stated, the pattern is stable across years and too large to be noise. It wants explaining.
The oldest answer is not about porn at all
Long before tube analytics existed, there was a documented, decades-old readership of women writing and reading erotic fiction about men with men. Slash fiction emerged out of media fandom in the 1970s, and the scholarship that grew up around it in the 1980s and 1990s established the basic demographic fact — the writers and readers were overwhelmingly women, most of them not gay men, and many of them straight.
That literature is worth knowing about because it kills the simplest explanation. This is not a recent phenomenon produced by streaming, algorithms, or the shape of the modern tube site. It predates all of them by forty years, in a medium with no visuals at all.
The explanations fan scholars reached for then were about narrative rather than bodies: two characters written as equals, neither of whom occupies the culturally scripted role of the person things are done to. That reading has held up better than most.
The arousal research, handled carefully
There is a well-established line of research, associated most closely with Meredith Chivers, finding that women's genital arousal responses are markedly less category-specific than men's — women's bodies respond to a broader range of sexual stimuli than their stated preferences would predict, and the correlation between physical response and reported desire is weaker in women than in men.
This gets misused constantly, so the caveats matter as much as the finding. Genital response is not desire, and the researchers say so explicitly. Non-specificity is not evidence of hidden orientation, and it does not mean a woman watching gay porn is secretly attracted to the scenario. What it does mean is that the folk model where arousal reliably tracks orientation is a poorer description of women than of men, and that a woman finding a scene she has no interest in participating in to be watchable is unremarkable rather than requiring special explanation.
What straight tube content optimises for
The production argument in the 2018 version of this article was crude but pointed at something real.
Mainstream straight tube content is shot for a presumed male viewer, and the camera shows it. Framing prioritises what that viewer wants to see, pacing is built around his arousal curve, and the woman on screen is lit and positioned as the object of the shot rather than as a participant with her own trajectory. None of that is a moral claim about the performers or the work; it is a description of a commercial optimisation that has been running for decades.
Gay male content is shot for an audience that wants to look at men, which means it is not solving that problem, which means a viewer who finds the straight format alienating is not alienated by it. That is a low bar, and clearing a low bar is often enough.
The feminist and women-directed corners of the industry exist and have for a long time, but they are a small share of what is free, indexed and one click away. Ease of access does a great deal of the work in any argument about what people watch.
Two explanations that should be retired
The 2018 article offered that older women watch gay porn to avoid comparing themselves unfavourably to a younger woman on screen, and that it functions as a safe outlet for a midlife crisis, since a gay young man can be fantasised about without the shame attached to fantasising about a straight one.
Neither has anything behind it. Both are assertions about the private motives of women who were not asked, both assume insecurity as the default female motivation, and the second one carries an implication about the age of the performers that is worth rejecting outright rather than reasoning about.
The absence-of-comparison idea does have a defensible version — that some viewers find scenes without a woman on screen easier to watch because there is no invitation to self-evaluate. That is a plausible account of an experience some people report. It is not the same as fear of a rival, and the difference is not cosmetic.
The useful takeaway
The interesting fact here was never that women like gay porn. It is that a genre can have an audience half composed of people it was never made for, sustain that for years, and have almost none of its production adjust in response.
Directors, marketers and platforms are still building for the audience they assume they have. If you want to look at either side of this yourself, the gay and lesbian sections are the places to start — and worth browsing with the knowledge that the label on the door is a poor guide to who is actually inside.