When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.

GPT-4 AI thinks Germany has the best Eurovision 2023 song entry

OpenAI logo on a GPT-4 graphic

Germany is a favourite to win this weekend’s Eurovision 2023 song contest, according to SongPad, a start-up which analysed each of the song entries using the latest OpenAI GPT model, GPT-4. The company used custom-designed prompts and parameters to review and rank each of the submitted songs and it decided Germany’s entry, Blood & Glitter by Lord of the Lost was best.

While GPT-4 may (or may not) have picked out the best song based on its quality, it’s important to remember that Eurovision can also get a bit political. Nobody will be surprised to hear that the winner of the competition last year was Ukraine and that Russia has been excluded due to its invasion of Ukraine. For this reason, the results may turn out different from what GPT-4 is predicting.

According to the little experiment from SongPad, the worst song submission is Serbia’s. What’s a little bit unfair in this analysis, though, is that English translations of songs were used before GPT-4 analysed the emotional impact, use of metaphor, rhyme, creative writing techniques, structure, and repetition. Unfortunately for Serbia, its song, Samo Mi Se Spava, was in English to begin with and it still got the worst rating from GPT-4.

The finals of Eurovision 2023 will take place on May 13 at 20:00 UTC+1 from the UK. The countries that will be taking part this year are Croatia, Czechia, Finland, France, Germany, Israel, Italy, Moldova, Norway, Portugal, Serbia, Spain, Sweden, Switzerland, Ukraine, and the United Kingdom.

Report a problem with article
Windows 11 Insider Preview
Next Article

Windows 11 (original release) Insider Release Preview Build 22000.2001 has many improvements

Previous Article

Windows 10 Insider Release Preview Build 19045.3030 includes improved search box experience

Join the conversation!

Login or Sign Up to read and post a comment.

4 Comments - Add comment