Wives Receive 2.3x More AI Love Songs Than Husbands, Neume Data Shows
-min.jpg){kind=small}
Abhinash Khatiwada
Men write AI love songs for their wives 2.3 times as often as women write them for their husbands, according to new data from Neume, an AI song generator. Across roughly 187,000 songs made on the platform, wives are the single most-named recipient, ahead of mothers, children and friends. The data suggests something that runs against a long pattern in how couples express affection: when an AI can supply the words, men are the ones reaching for it.
- Songs for a wife outnumber songs for a husband 2.3 to 1.
- Wives are named 2.5 times as often as mothers, the next most common recipient after husbands.
- The gap is widest on birthdays (3.4x) and peaks in February (3.8x). June, Father's Day month, is the only month it reverses.
- Prompts written for a wife are about 30% shorter. Men give the AI a sketch; women spell out what they want to say.
- Research on how men are taught to handle emotion offers one explanation: many men feel attachment but struggle to put it into words.
Wives are the #1 recipient of AI love songs on Neume
Neume looked at every song created between February 2025 and September 2026 where the prompt explicitly named who the song was for: "for my wife," "to my husband," "about my mom" and their common variants. Counting unique users, and comparing each recipient with husbands, the ranking looks like this.
| Recipient | Compared with husbands |
|---|---|
| Wife | 2.3x |
| Husband | 1.0x |
| Mother | 0.9x |
| Daughter | 0.9x |
| Father | 0.8x |
| Girlfriend | 0.8x |
| Son | 0.6x |
| Boyfriend | 0.6x |
| Best friend | 0.6x |
| Ex | 0.3x |
Three things stand out. Wives lead every other recipient, including mothers. Parents are close to even, with mothers named only about 1.1 times as often as fathers. And the romantic gap is small before marriage and large after it: girlfriends outnumber boyfriends about 1.3 to 1, while wives outnumber husbands 2.3 to 1.
Why this is surprising: women usually do the card work
Expressing affection in writing has long been treated as women's work. According to the Greeting Card Association, women buy more than 80% of all greeting cards and spend more time choosing a card than men do.
Researchers have a name for this. In 1987 the anthropologist Micaela di Leonardo described kin work: the conception, maintenance and celebration of family ties through visits, letters, cards and holidays, work done mostly by women. Her paper's title says it all: "The Female World of Cards and Holidays." Sociologist Carolyn Rosenthal had described the same role in 1985 as "kinkeeping", and more recent research in Sex Roles continues to study it.
On Neume, that pattern flips. When the message is a song and an AI helps write it, the spouse doing the expressing is more often the husband.
The gap holds on every occasion
The wife-first pattern is not driven by one holiday. When a prompt names both the spouse and the occasion, wives lead every time.
| Occasion | Songs for a wife per song for a husband |
|---|---|
| Birthday | 3.4x |
| Anniversary | 2.7x |
| Valentine's Day | 1.7x |
Birthdays show the widest gap. A birthday comes every year, needs something new to say every year, and has no shared script the way Valentine's Day does.
Romance on a deadline: the February spike and the June reversal
The calendar tells its own story. Wife songs spike in February, Valentine's month, when users writing for a wife outnumber those writing for a husband roughly 3.8 to 1. Mother's Day month shows the same lean, at about 3 to 1. Then June arrives, the month of Father's Day, and it is the only month of the year when husbands receive more songs than wives.
Read the monthly view as a pattern rather than precise ratios: October to December data comes from 2025 only, and nearly all of the February data comes from 2026. The shape is still consistent. Songs for wives cluster around dates that demand a romantic gesture.
Men say less: shorter prompts, more left to the AI
The prompts themselves show a second difference. People writing for a wife write less.
| Metric | For my wife | For my husband |
|---|---|---|
| Median prompt length | 339 characters | 476 characters |
| Prompts under 200 characters | 36.5% | 26.2% |
| Mentions "thank you" or "grateful" | 14.4% | 14.6% |
The median prompt for a wife is 339 characters, against 476 for a husband: about 30% shorter. More than a third of prompts written for a wife are under 200 characters. Gratitude shows up at the same rate on both sides, so this is not about feeling less thankful. The difference is in how much of the message people write themselves. Prompts for a husband tend to spell out what the writer wants to say. Prompts for a wife more often hand the AI a sketch and let it find the words.
Why men reach for the words: normative male alexithymia
Psychology offers a plausible explanation. The psychologist Ronald Levant describes normative male alexithymia: a socialized difficulty many men have with identifying, describing and, most of all, expressing emotions. In Levant and Parent's account, boys are raised to hold back two kinds of emotion in particular: vulnerable ones and attachment ones, the feelings of caring for and needing someone. Discouraged from talking about them, many never build a vocabulary for them.
Those attachment emotions are exactly what a love song is made of. As a Simply Psychology explainer on this research puts it, physiological studies show men's bodies respond to emotional stimuli as strongly as women's, while their self-reports and facial expressions show less. The gap is in labeling and expressing emotion, not in feeling it. Read that way, the Neume data does not show men caring less. It suggests men who care a great deal, and who find a tool that turns a few plain words into something that sounds the way they feel.
It is not just Neume: men say they plan to use AI for love notes
Survey data points the same way. In a 2023 McAfee study of more than 5,000 adults in nine countries, 30% of men (26% of all adults) said they planned to use AI to write a Valentine's love letter. The most common reason was that it would make them feel more confident (27%).
A year later, McAfee's 2024 survey of 7,000 adults in seven countries found 45% of men (39% of all adults) considering it. Surveys capture what people say they intend to do. The Neume data shows what people actually do when the tool is in front of them.
Two objections worth taking seriously
"Men just use AI more"
They do. A Harvard Business School working paper pooling surveys of 318,924 respondents estimated generative AI adoption at 47.8% for men and 39.3% for women, a relative gap of 22%, which has since settled around 16%. That is a gap of about 1.2 to 1. The wife versus husband gap on Neume is 2.3 to 1, and about 3.8 to 1 in February. General AI adoption explains only a small part of it.
"But men say 'I love you' first"
They do that too. Across six studies, Ackerman, Griskevicius and Li (2011) found that although people believe women confess love first, it is actually men who say it first, and a 2022 study by Watkins and colleagues found the same pattern across several countries.
That is not a contradiction. It sharpens the finding. Saying three words early in a relationship is one thing. Saying something new and specific on the fifth birthday, the tenth anniversary, the twentieth February, is another. The Neume data is almost entirely about wives, not new partners. It suggests men are quick to declare love and quieter at the ongoing, elaborated expression of it years into a marriage. That is where the AI steps in.
The tension in outsourcing love
There is an irony here. The same McAfee surveys that found men planning to use AI also found that people do not like being on the receiving end: 49% in 2023 said they would be offended to learn a love note was written by a machine, and 57% in 2024 said they would be hurt or offended to find out AI wrote their Valentine's message.
A song may be a different case from a note. Nobody expects a spouse to write, sing and produce a full track alone, and the details that make it land (the name, the first apartment, the inside joke) still come from the person asking. If there is a lesson in the data, it may be this: the AI can supply the melody and the phrasing, but the specific memory has to come from you. Men who put more of themselves into the prompt get a song that sounds more like them.
If you want to try, the song for my wife and song for my husband generators start from exactly those details, and our guide to making a personalized song gift covers what to include. For another look at what people make on Neume, see what 123,000 AI-generated songs reveal about AI music.
Methodology
Neume analyzed anonymized text prompts from 186,769 songs created on the platform between February 2025 and September 2026, made by 44,941 users on iOS, Android and the web; 97% of songs had a user-written prompt. Recipients were identified from explicit phrases such as "for my wife," "to my husband" and "about my mom," including common variants (wifey, hubby, mum) and adjectives ("for my beautiful wife"). Comparisons are based on unique users. Neume does not collect user gender, so the writer's gender is inferred from the recipient named; this is a strong proxy but not perfect, since it includes same-sex couples.
Limitations. Songs that name a spouse are a small share of the full dataset, so the ratios carry more uncertainty than the overall totals. Many users never name a recipient, so these figures describe only people who did.
Men are not saying less because they feel less
The data suggests a simple reading. Women have long carried the work of putting affection into words. Give men a tool that finds the words for them, and they show up, especially for their wives, especially on a deadline. The feeling was there. What changed is that the words got easier to find.
Frequently Asked Questions
On Neume, the data suggests men do, at least for spouses. Songs written for a wife outnumber songs written for a husband 2.3 to 1, and wives are the most-named recipient on the platform. Neume does not collect gender, so this is inferred from the recipient named, which includes same-sex couples.
One explanation comes from research on normative male alexithymia, described by psychologist Ronald Levant: many men are socialized to hold back vulnerable and attachment emotions and never develop words for them. The feeling is there; the vocabulary is harder. An AI can turn a few plain details into something that sounds the way they feel.
They do, but not by enough to explain this. A Harvard Business School working paper estimated generative AI adoption at 47.8% for men and 39.3% for women, about 1.2 to 1. The wife versus husband gap on Neume is 2.3 to 1, and about 3.8 to 1 in February.
February, Valentine's month, is the peak, with wife songs outnumbering husband songs roughly 3.8 to 1. June, the month of Father's Day, is the only month where husbands receive more songs. Read the monthly view as a pattern rather than a precise ratio.
Some might. McAfee surveys found 49% of people in 2023 and 57% in 2024 said they would be offended or hurt to learn AI wrote a love note. A song is different from a note, and the personal details still come from the sender, so the more specific memories you put in, the more the song sounds like you.
Neume analyzed anonymized prompts from 186,769 songs made between February 2025 and September 2026 by 44,941 users. Recipients were identified from explicit phrases like “for my wife” and “to my husband,” including common variants. Comparisons are based on unique users.
Find the words for your person
Tell Neume a few real details about them, a name, a memory, a place, and get a full song with vocals in about a minute. Your first High Quality song is free.