Here's what 'LOL' looks like in ancient Egyptian hieroglyphics

A new AI-powered tool from Google decodes ancient Egyptian hieroglyphics, and even lets you translate modern words and emojis into the 5,000-year-old symbols. 

An "LOL," I learned, would have taken a lot more characters back in the day of the pharaohs -- a string of 12 that includes images of a couple of birds and a lower leg, according to the platform. And the face-palm emoji looks like a guy head-planting, or something. 

World Emoji Day Friday might have just gotten a lot more interesting. 

The Google Arts & Culture team released the tool -- called Fabricius -- on Wednesday, the anniversary of the discovery of the famed Rosetta Stone. The slab from 196 BCE bears writing in three different scripts and is considered key to deciphering Egyptian hieroglyphs. 

Fabricius is an interactive tool with three sections: "Learn" takes you through the basics of hieroglyphics. "Work" lets you upload pictures of real-life symbols from temple walls and artifacts. And "Play" is where you get to fuss around with words, phrases and emojis to see what your text message would have looked like back when tablets didn't have power buttons. You can share the messages on social media or over email, though Google warns that the hieroglyphs won't always reflect accurate translations of your messages. 

"This experiment explores the potential for using machine learning to increase efficiencies in the translation of ancient languages and open new avenues for academic research," the Fabricius site reads. Google calls it the first digital tool for translating and deciphering hieroglyphics. 

Google created Fabricius in collaboration with the Australian Centre of Egyptology at Macquarie University, Psycle Interactive and video game company Ubisoft, which previously researched  hieroglyphics to coincide with the release of Assassin's Creed Origins. That action adventure game is set in ancient Egypt. 

Google used Google Cloud's AutoML, a suite of machine learning products, to create an AI model able to make sense of hieroglyphs. But one Egyptologist said machine learning can't yet replace humans when it comes to understanding hieroglyphics' context and variants.  

"While impressive, it is not yet at the point where it replaces the need for a highly trained expert in reading ancient inscriptions," Roland Enmarch, a senior lecturer in Egyptology, at the University of Liverpool, told BBC News. "There remain some very big obstacles to reading hieroglyphs, because they are handcrafted and vary enormously over time in level of pictorial detail and between individual carvers/painters." 

Fabricius is being released as open source to support further research in the study of ancient languages. In the meantime, translating smiley faces and winkies into hieroglyphics is a colorful exercise in linguistic history. Good fortune upon you as you play around with it.