If you’re a fan of the spy genre, then you’ll know that sending coded messages is an essential way of getting messages across to operatives without giving away the secret. Recently, scientists have allowed encryption of messages that essentially let you send secret messages in plain sight – using Helvetica and Times New Roman.
That’s right: fonts are now the new encryption device. Researchers at Columbia University have figured out a way to encrypt messages in these fonts by making tiny changes to them – too small for the human eye to pick up, but it’s no problem for a computer vision algorithm. That technology is dubbed Fontcode.
In a demo, they showed a secret message – “Greetings, my friend! You finally found me.” – into a paragraph taken from The Lord of the Rings.
How does it work?
The algorithm slowly shifts letterforms of one typeface to make tiny changes in the shape of every letter that the human eye can’t detect. So an “h” can be slightly thicker in the stem, or the curve of a “j” slightly sharper.
The researchers made 52 variations of each letter, each of which corresponds to every other lowercase and capital letter in the alphabet (and theoretically every numeral and punctuation mark as well). These 52 variations then go into a “code book” that the computer uses to match the altered letter with the secret letter it’s encoding.
This code book and machine learning algorithm can help ensure that documents haven’t been tampered with – by embedding an authentication code or secret message into the font itself, someone can use the decoding algorithm to verify the document hasn’t been altered if the message is still there.
How’s it going to be used?
According to Changxi Zheng, an associate professor in computer science at Columbia who led the research, says that companies in the legal space and intelligence agencies have already reached out with interest.
However, Fontspace can also apply to our daily lives the same way QR codes or barcodes do. For example, if you saw an ad on a wall, you can simply snap an image of the whole thing to reveal more information. For designers, it can make messages neater.
You could also embed metadata into documents without altering how the document looks – kind of like how metadata is embedded in photos already. So your textbooks can be a portal to something more, for instance.
Right now, the tech is still in developmental stage, but who knows when it’ll be unleashed into the world?
