Men are using AI to write love letters—and people hate it

If you're struggling to find the right words for your Valentine's Day card this year, you're definitely not alone. It turns out a massive chunk of the population is just outsourcing the whole thing to ChatGPT.

According to McAfee’s latest "Modern Love" study, which surveyed 7,000 people across seven countries, nearly half (45%) of all men plan to use artificial intelligence to write their Valentine's messages this year. That is a noticeable jump from 30% last year. Across all American adults surveyed, 39% are planning or considering using AI to draft romantic notes, up from 26% a year ago.

It works surprisingly well, until you get caught

People aren't just doing this out of laziness; they are doing it because the results speak for themselves.

Nearly seven in ten respondents (69%) admitted that using AI-generated content actually helped them get more interest and better responses than their own original writing. Around 30% of men and 27% of women reported using AI to polish their dating profiles, touch up photos, and smooth out chat messages.

Here is the catch: nobody actually wants to be on the receiving end of code.

A solid 57% of respondents stated they would be hurt or deeply offended if they found out their Valentine's Day message was hallucinated by a bot. On top of that, 64% expressed distrust toward potential matches who use AI-generated imagery or chatbots on their profiles. The real kicker is that 46% of Americans admit they probably can't even tell if a love message was written by an algorithm or a human.

The romance scam machine is scaling up

While everyday daters are using AI as a harmless wingman, cybercriminals are using the exact same technology to supercharge romance scams.

Roughly 31% of Americans said they have communicated with an online love interest who turned out to be a scammer. More than half (57%) of respondents reported being asked to send money shortly after meeting someone online, with 30% of those targets being asked for over $1,000. Others were pressed for passwords, birthdays, intimate media, and even Social Security numbers.

McAfee Labs reported a 25% surge in Valentine-related malware campaigns, a 300% increase in malicious URLs, and an astounding 400% spike in romance-themed spam and phishing campaigns leading up to the holiday.

Daters are adapting by turning into amateur private investigators. About 38% now use reverse image searches on profile pictures, and 59% regularly dig through social media to vet potential partners. For 13% of those background-checkers, the sleuthing is what revealed they were talking to an outright scammer.

Why it matters

We think this shift exposes a fascinating paradox in consumer AI: generative tools are creating an inflation of charm where polished words are becoming practically worthless.

For daters, running your thoughts through an LLM to fix a typo is fine, but letting a model generate your entire romantic sentiment is a terrible gamble. The whole currency of romance is effort and vulnerability. If your partner figures out you copied and pasted a prompt output, the emotional damage far outweighs any points you scored for poetic prose.

For dating apps and social platforms, the stakes are existential. Generative AI makes high-effort romance scams frictionless to automate at massive scale. The bots can now hold coherent, emotionally resonant conversations indefinitely. As distrust grows, platforms will soon have to shift their entire value proposition away from matchmaking algorithms and toward robust proof-of-humanity verification.

Ultimately, when perfect, algorithmically optimized flirting becomes the default, authentic human awkwardness might end up being the only real proof of genuine affection left.