Zendoric
← Back to the day · October 15, 2026

When a university website becomes bait: 'AI girlfriend' spam abuses legitimate links

🕒 Published on Zendoric: October 15, 2026 · 00:13

✨ AI-generated · how it's made

A headline about 'AI girlfriend premium vs free' arrives under the University of Córdoba's domain, but there is no news: just a link with suspicious parameters. What's interesting is the technique, not the product.

🎧 Listen to the analysis · synthetic voice

By University of Córdoba (UCO) · October 14, 2026.

This material contains no news. All that comes through is an ad-style headline ("AI Girlfriend Site Premium Vs Free AI Companion First Month Discount 2026") and a description that repeats it. There are no figures, no specific product and no facts to report. That's why this piece is short on purpose.

What can be observed is the shape of the URL. It sits on an institutional domain (uco.es) but carries very long parameters. These include a base64-encoded block (base64 is a format that turns data into text) and a "video" identifier stuffed with commercial keywords. Judging by the link itself, one of those parameters appears to point to a file hosted on an external third-party service. We have not opened or verified that content, so we make no claims about what it does. Nor can we say the university is behind anything. Most likely, someone is exploiting a legitimate page to make a promotional link look trustworthy.

Our take: this is a well-known malicious SEO pattern. A page belonging to a reputable institution is used to rank links in lucrative niches, such as AI virtual girlfriends. The bet is that both the search engine and the reader will lower their guard because of the domain.

More broadly, the AI companion sector brings in a lot of money through subscriptions and "first month" discounts, and that attracts legitimate services and scams alike. The practical rule is simple: a respectable domain does not vouch for where a link with odd parameters ends up. When you see a deal-style headline on an academic website, the prudent move is not to click.

In the long run, the episode illustrates a near-term risk that AI doesn't create but does make cheaper: the mass production of believable lures. The defense lies in better filters in search engines and aggregators, and in readers who grow suspicious when the form doesn't match the content.

🔗 Related on Zendoric

Sources & references