This is the best summary I could come up with:

“As we investigated available CAPTCHA options, we weren’t satisfied, so we decided to develop our own,” Eamonn Maguire, a former Facebook engineer who now heads up Proton’s machine learning team, wrote in a blog post.

This is usually presented to the user in the form of a visual or cognitive challenge, one that is relatively easy for a human to complete but difficult for a machine.

CAPTCHAs, while generally effective, come with trade-offs in terms of usability, accessibility, cultural biases, and annoyances that businesses would prefer not to impose on their users.

This is why companies such as Apple and Cloudflare have sought ways to tell the difference between humans and bots automatically using alternative mechanisms, such as through device and telemetry data.

And while there are other alternative CAPTCHA services out there, given Proton’s core raison d’être, it clearly does make sense to develop its own — as resource-intensive as that may be.

“In this manner, a botnet that can bypass the initial proof of work but struggles with the visual challenges will be met with increasingly complex computations.

The original article contains 642 words, the summary contains 179 words. Saved 72%. I'm a bot and I'm open source!