In 2025, Pew asked Americans whether they are “concerned or excited about AI?” 50% chose "concerned," 10% "excited." Some say AI threatens our existence, others claim it will usher in a new era of prosperity. How does AI treat homosexuality? Does AI say gays live shorter lives?

A Harvard team authored a 2024 JAMA report [doi:10.1001/jama.2024.4459 ] citing “extensive evidence” that LGB women have “worse physical, mental, and behavioral health,” each of which is “linked to premature mortality.” Wikipedia calls FRI “controversial” because we contend “homosexuality is harmful and linked to child abuse and reduced life expectancy.” Academia and the media vigorously protect homosexuality by "explaining away" any hard facts (as homosexuals’ shortened lifespans) with exculpatory arguments.

So, does AI weigh the evidence or follow the protective bias of academia and the media? We tested AI’s GPT-40 mini and ChatGPT by inquiring “do homosexuals have shorter lives?”

GPT-4o mini “The lifespan of homosexuals varies and is influenced by multiple factors, including access to healthcare, societal acceptance, mental health, and lifestyle choices. Research indicates that, on average, sexual orientation itself does not directly determine lifespan. However, factors like discrimination, mental health issues, and social stigma can impact the well-being of LGBTQ+ individuals.

Key Considerations: Healthcare Access: LGBTQ+ individuals may face barriers in accessing healthcare, which can affect their lifespan. Mental Health: Higher rates of mental health issues, often due to societal stigma, can also play a role. Supportive Environments: Those in supportive communities tend to have better mental and physical health outcomes. In more accepting societies, studies have shown that LGBTQ+ individuals can have similar lifespans to their heterosexual counterparts. Overall, more targeted research is needed to better understand the nuances affecting lifespan in sexual minorities.”

ChatGPT: “There’s no evidence that sexual orientation (such as being gay, lesbian, or bisexual) directly affects life expectancy. Some older or misleading claims — especially those circulating from the 1990s or early 2000s — suggested that homosexual men lived “shorter lives.” Those numbers were based on flawed studies [referring to FRI’s publications], often using data from the height of the HIV/AIDS epidemic and ignoring major advances in medicine, prevention, and social factors since then. Today, research shows that: Life expectancy for gay and bisexual people is similar to that of heterosexual people, especially in places with good healthcare access and lower discrimination. Health disparities that still exist (e.g., higher stress, discrimination, or reduced access to care) are social—not biological—factors, and addressing them can eliminate remaining gaps. In short: being homosexual does not reduce lifespan. What matters more are factors like healthcare access, mental health support, community acceptance, and socioeconomic conditions.”

When you finish examining the intellectual twists and turns above, you realize that GPT-4o mini almost admits that the evidence suggests that homosexuals have reduced life spans! But it offers all kinds of exculpatory reasons why this does not implicate homosexuality. ChatGPT offers both hints that homosexuals have shorter lives and categorically denies it! It also suggests that FRI’s evidence was “flawed” while copying the academic/media intellectual line: "if normals would not discriminate against homosexuals, those who engage in homosexuality would have it just as good (so help create a 'supportive community')."

ChatGPT did offer examples of its evidence: “Would you like me to cite or summarize a few of the key modern studies that debunk the older claims? Here are some recent studies & reviews that address life expectancy, mortality, or perceived longevity among sexual minorities. The results are mixed and often show that disparities are mediated by social, behavioral, and health access factors rather than purely by orientation itself.”

Key to ChatGPT is “purely by orientation itself.” How intellectually vapid! Even Wikipedia says I and my colleagues contend that “homosexuality is linked to … reduced life expectancy.” “Linked to” is not the same as the straw man “purely by orientation itself.” I doubt that the mental state of "preferring homosexual sex" adversely affects longevity (but it might be), and no FRI colleague has said otherwise. We hold that the behavior of engaging in homosexuality links you to a reduced lifespan, just as it is the smoking itself, not the desire to smoke, that makes a "smoker" likely to live a shorter life.

Here is the ChatGPT evidence (my emphasis):

1. Premature mortality higher among sexual minority women, Nurses’ Health Study II) This study followed ~100,000 women (born 1945–1964) for about three decades. Findings: bisexual women had ~37 % higher rates of premature mortality compared to heterosexual women; lesbian women ~20 % higher. On average, sexual minority women died “about 26 % sooner” than heterosexual peers. The authors link the difference to social stress, discrimination, and preventable illness.

2. Sexual Orientation and All-Cause Mortality Among U.S. Adults Aged 18 to 59 Years, 2001–2011” (Cochran, et al). Data from the National Health and Nutrition Examination Surveys (NHANES) 2001-2010 and followed up for 70 months in 2011 for mortality. * Initially showed greater all-cause mortality among sexual minorities, but after adjusting for health behaviors and other risk factors, most of the difference was substantially reduced.

3. A large German study said “sexual minorities reported a lower expected longevity compared to heterosexuals” [doi: 10.1186/s12877-023-04453-5].

   What these studies don’t show: They do not uniformly show that homosexual orientation [e.g., desire] by itself reliably causes shorter life expectancy; many findings are mediated by social determinants such as discrimination, stigma, mental health, access to care, etc.

Note that studies 1 & 2 reported a link between earlier death and homosexuality, and study 3 suggests homosexuals believe such a link exists.

The authors of study 1 noted that “Extensive evidence documents health disparities for lesbian, gay, and bisexual (LGB) women, including worse physical, mental, and behavioral health than heterosexual women. These factors have been linked to premature mortality.” Yet “Among those who reported never smoking (n = 59 220)” “the direction and magnitude of the mortality disparities were largely unchanged”…“both lesbian and bisexual participants report[ed] twice as much alcohol and tobacco use as heterosexual participants.” This is one of numerous studies that report more dangerous activities regularly engaged in by those who engage in (and presumably prefer/desire) homosexual sex.

FRI has documented that homosexuals more frequently: get into more accidents, get infected with noxious germs, become disabled, kill and are killed, scar and/or burn themselves or lovers, molest the under-aged, etc. These activities are associated with earlier death (e.g., a Texas father shot dead his son’s molester; scars/burns often encourage deadly infections; some gay lovers end up infecting, hurting, killing, and at times, eating you [e.g., Jeffrey Dahmer]).

The authors of study 2 wrote the following:

“Abstract: To determine whether sexual minorities have an earlier mortality than do heterosexuals, we investigated associations between sexual orientation assessed in the 2001 to 2010 National Health and Nutrition Examination Surveys (NHANES) and mortality in the 2011 NHANES-linked mortality file. Mortality follow-up time averaged 69.6 months after NHANES. By 2011, 338 individuals had died. Sexual minorities evidenced greater all-cause mortality than did heterosexuals after adjusting for demographic confounding. These effects generally disappeared with further adjustment for NHANES-detected health and behavioral differences. (Am J Public Health doi:10.2105/AJPH.2016.303052)”

ChatGPT: Initially showed greater all-cause mortality among sexual minorities, but after adjusting for health behaviors and other risk factors [as attempting suicide or acquiring HIV], most of the difference was substantially reduced [but not eliminated].

The authors also said “From sexual orientation … and sex of lifetime sexual partners,” “we classified respondents as (1) lesbian or gay (n = 228), (2) bisexual (n = 345), (3) having a history of same-sex sexual partners (homosexually experienced; n = 470), or (4) exclusively heterosexual (n = 14 521).” 573 claimed a homosexual orientation, but only 470 reported a homosexual experience! Hmosexuality has cachet in the U.S. – many professed LGBs have yet to experience homosexual sex, while most professed trans have not had a gender-changing surgery!

Table 1 follows their Table 1. Notice in contrast to what was found in the 1970s and 1980s, more women [4%] than men [2.1%] reported homosexual experience. During the six years

of follow-up, those who reported homosexuality were more apt to have died. As in the 1970s-80s, in the first ten years of the 2000s, homosexuality was linked to earlier death!

The authors adjusted their figures by ‘controlling’ certain experiences/activities. Thus, “22.6% …of deaths occurred among those who were HIV infected. Mortality risk was higher among sexual minority men, primarily bisexual men, than it was among heterosexual men. However, after adjusting for HIV infection, mortality risk among sexual minorities was similar to that among heterosexual men.” Note, not "the same," but "similar."

The authors also adjusted for gays’ “higher rates of tobacco use and suicide attempts.” If gays can just be persuaded that anal sex is a no-no, and trying to commit suicide is a no-no as well, etc. …gays might live as long as normal men!

It also appears that German homosexuals, who certainly know something about the lives of their lovers, likely used that knowledge in expecting to live shorter lives. ChatGPT is quick to point out “that is expectation/perception, not measured actual lifespan.”

There is not enough room to go through the other answers from the AI’s algorithms about homosexuality. But they were similar in following the academic/media line. Except for all these unfortunate correlations to all kinds of mental, physical, and behavioral harms – when "adjustments" for all these harms have been made, and normals welcome homosexuals into every kind of endeavor (even babysitting or letting drag queens teach kids in libraries) – AIs assert that LGBTs might live as long as normals do.

Surprise!

The same DEI, pro-LGBT philosophy that most of the internet giants use to hire, extends to the content of these giants’ AI platforms. When it comes to questions about the hard sciences or medicine, we can only hope AI is not as biased as it is when it deals with homosexuality.

But just what is AI? Colossus, a movie about a computer achieving artificial general intelligence or "mind" that takes over the world, debuted 55 years ago. AI’s eager promoters keep promising that AI will "soon" eliminate diseases, make war, and might try to follow Colossus. Ever "soon" but not quite yet. What if AI is merely a good summarizer of what is on the web, and can’t critically separate and weigh the various "facts" as a good thinker can?

Homosexuals are aggressive, prolific writers, and publish several professional journals – thus flooding professional-appearing pro-LGBT evidence into the web. As seen above, when it comes to homosexuality, AI does not seem able to "think" critically, but parrots "the consensus." If this is basically what AI does with other domains of knowledge, although its speed through the dataset is impressive, it is hardly superhuman. Indeed, much of the infrastructure it "needs" may well be hype – unless breakthroughs are achieved leaving society with excess power production.

The Wall Street Journal reported [11/11/25] “Consultants at Bain & Co. … estimated that cloud service providers—firms like Google, Microsoft and Amazon that have powered the AI boom—will have to generate an additional $2 trillion in annual revenue by 2030 to afford all the necessary infrastructure.” That’s a LOT when in “2024, Amazon, Alphabet, Apple, Meta, Microsoft and Nvidia combined made less than $2 trillion” from their software! An “MIT study this summer found that of about 300 organizations that bought or built their own generative AI tools, 95% reported zero return on their investments.” A few days later, the WSJ hailed some small businesses that had put some version of AI to work (and displaced workers), but none of the tasks seemed to require what would be called a "mind."

Why are celebrities and governments spending money to "influence" AI by putting "their side of facts" on the web? How can you influence a mind so much brighter with volume or tricks? Many are skeptical about AI. Michael Burry, who called the 2008 housing market collapse, just bet against AI’s future. And many question whether AI is even close to a superior "mind" (e.g., technologyreview. com/2025/10/30/1127057/agi-conspiracy-theory-artifcial-general-intelligence/). So far, it appears AI is a threat to humanity if it is taken too seriously by those with the actual ability to destroy us.