It's not your imagination. CAPTCHA tests are getting harder (2024)

SCOTT SIMON, HOST:

Just sec here. I'm trying to prove I'm not a robot. Which squares have trees? Wait, is that a tree or a bush? Or Is that BJ Leiderman, who does our theme music? In fact, CAPTCHA tests are getting harder to answer. Amanda Fennell, chief information security officer for Prove, a cybersecurity firm, and an adjunct professor at Tulane University, joins us. Thanks so much for being with us.

AMANDA FENNELL: Yeah. I'm glad to be here.

SIMON: First, with all respect and courtesy, are you a robot?

FENNELL: I might be. We're going to have to do a couple of tests to find out.

SIMON: (Laughter) You took my next question. Now, CAPTCHA stands, I'm told, for Completely Automated Public Turing test to tell computers and Humans Apart.

FENNELL: That is correct.

SIMON: That very name is chilling.

FENNELL: Well, the name actually kind of came before the test a little bit. They knew they needed to do something, and there were a lot of really smart people out there who were working on it.

SIMON: So how did they develop - who said, I know what we can do?

FENNELL: Well, there were three researchers at the time who were on kind of the forefront of machine learning and decided - pretty much since computers have come out, it's like "Blade Runner." The next thing they said was, how do we prove it's not a computer? And so that was pretty much from the beginning. Started in probably about the '90s. There was a patent put out in '97, and then these three researchers kind of coined the term in 2003.

SIMON: Why are CAPTCHA tests getting harder?

FENNELL: Well, because computers are getting smarter, probably. So we're trying - we're just doing the constant arms race, if you will. The computers get smarter because humans are getting, you know, more applicable with their computers, so they're just trying to build a better mousetrap.

SIMON: What are some of the hardest CAPTCHA tests out there that you've seen?

FENNELL: Well, I actually am one of those people who's challenged by them. There's a percentage of the human population, about 3%, actually, that have a literal issue whenever they see these kind of stimulant tests. And so, for me, personally, they're horrible. And it doesn't matter if I do the audio or the visual. I have a high probability of failing it.

SIMON: And you're a cybersecurity expert.

FENNELL: I know. That's what they say. Yeah. But there are some better alternatives to what they've been using for CAPTCHA, the version 1, 2, 3, ReCAPTCHA with Google. There are better ideas that are coming out in more recent years.

SIMON: And they are coming out, right? There are things on the horizon?

FENNELL: You may have seen some of them. My personal favorite is actually gamification. It's, you know, some kind of an image that'll say, can you plant a garden? And you have to move the images that make sense. Simple questions, you know, what's one plus one? Things like that. Sliders, which all of us Apple users love to see when we see a slider across the screen. But a lot of things are actually happening behind the scenes. This is actually concerning. And I don't want to get on a soapbox, but this...

SIMON: Yeah.

FENNELL: ...Is the concern with reCAPTCHA. Google acquired reCAPTCHA in 2009, and it is the behind-the-scenes way of doing CAPTCHA where the human does not have to interact, and it takes all the knowledge of your internet browsing. So they actually have a lot of data privacy concerns, because how else are they proving you're a human? They're using behavioral analysis of your internet search.

SIMON: Oh, my gosh.

FENNELL: I know. I don't mean to scare everyone, but yeah, it's a thing that you just have to be in front of, understand, do some private web browsing, don't allow cookies, things like that - just some healthy hygiene.

SIMON: But, I mean, as chilling as some of us might find that, would history suggest that if it means we can get our toothpaste delivered in eight hours, we'll do it?

FENNELL: Right. And that's really the problem we've been having with CAPTCHA, is that they're trying to be smarter about deterring bots, but they're trying to make it also a frictionless experience for humans, and that's not working, which is why people have found these alternatives.

SIMON: Amanda Fennell is chief information security officer for Prove, a cybersecurity firm, and adjunct professor at Tulane University. Thanks so much for being with us, and may all your CAPTCHAs be easy to solve.

FENNELL: That was adorable.

(LAUGHTER)

FENNELL: I appreciate you having me. This was a lot of fun.

SIMON: And we should note Google is a funder of NPR, but we cover them like we cover everybody and anybody else. Transcript provided by NPR, Copyright NPR.

NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.

It's not your imagination. CAPTCHA tests are getting harder (2024)

FAQs

Why can't robots pass CAPTCHA tests? ›

The letters are distorted so that bots are not likely to be able to identify them. To pass the test, users have to interpret the distorted text, type the correct letters into a form field, and submit the form.

What is the test to tell Computers and humans apart? ›

CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) is a type of security measure known as challenge-response authentication.

Are you a human CAPTCHA? ›

CAPTCHA describes a spam protection method whose goal is to distinguish automatically generated input from human input and filter it out accordingly. CAPTCHA is an acronym and stands for “Completely Automated Public Turing test to tell Computers and Humans Apart”.

Why can't AI beat CAPTCHA? ›

In summary, captchas are intentionally designed to be a barrier to automated bots and a means of distinguishing humans from machines. While technology continues to advance, the constant evolution and complexity of captchas make it an ongoing challenge for robots to pass these tests with the same ease as humans.

Can AI outsmart CAPTCHA? ›

Using neural networks trained on massive datasets, AI can now reliably solve CAPTCHAs designed to leverage distortions or clutter to stump bots. Whereas older machine learning approaches struggled with obscured or stylized text, modern AI has achieved human-comparable accuracy on many CAPTCHA datasets.

Has any AI passed the Turing test? ›

The program, known as Eugene Goostman, is the first artificial intelligence to pass the test, originally developed by 20th-century mathematician Alan Turing. The machine was tasked with persuading 30 human interrogators of its humanity, communicating with them via a series of five-minute keyboard conversations.

Did ChatGPT pass the Turing test? ›

ChatGPT passes the famous 'Turing test' - suggesting the AI bot has intelligence equivalent to a human, scientists claim.

What did Alan Turing say about AI? ›

Turing gave quite possibly the earliest public lecture (London, 1947) to mention computer intelligence, saying, “What we want is a machine that can learn from experience,” and that the “possibility of letting the machine alter its own instructions provides the mechanism for this.”

What is the paradox of CAPTCHA? ›

Notice the paradox: a CAPTCHA is a program that can generate and grade tests that it itself cannot pass (much like some professors).

Why is Google asking if I'm human? ›

When you see a message from Google or encounter a page that requires you to respond to a CAPTCHA before you proceed, it's because Google is out to foil hacking and automated spamming programs, also called "bots." Unusual network activity, such as numerous messages from your computer, may trigger a Google message ...

Why does Google ask I am not a robot? ›

reCAPTCHA is a free service from Google that helps protect websites from spam and abuse. A “CAPTCHA” is a turing test to tell human and bots apart. It is easy for humans to solve, but hard for “bots” and other malicious software to figure out.

Why can't CAPTCHA be automated? ›

Captchas are mostly dynamic and are intended to defeat automation frameworks like Selenium. However, since they are dynamic, they can be carried out successfully only with human interpretation and analysis, and as a result, most of the automated bots fail.

Why are robot CAPTCHAs so hard? ›

Bots attempt to mimic humans, but they're evolving to be faster and better at outsmarting the barriers to entry online — so CAPTCHA creators are crafting ridiculously difficult prompts to keep automated hackers, spammers and site crashers at bay.

Can a robot bypass CAPTCHA? ›

Image Recognition reCAPTCHA Test

The image recognition test is fairly easy for modern bots to bypass. Modern AI integrated with bots allows malicious automation to recognize images and pass the test as a human would.

Can AI pass CAPTCHA test? ›

In light of these developments, it's no surprise that recent research has found that AI bots are now outperforming humans in solving CAPTCHAs.

Top Articles
Latest Posts
Article information

Author: Kerri Lueilwitz

Last Updated:

Views: 6539

Rating: 4.7 / 5 (47 voted)

Reviews: 94% of readers found this page helpful

Author information

Name: Kerri Lueilwitz

Birthday: 1992-10-31

Address: Suite 878 3699 Chantelle Roads, Colebury, NC 68599

Phone: +6111989609516

Job: Chief Farming Manager

Hobby: Mycology, Stone skipping, Dowsing, Whittling, Taxidermy, Sand art, Roller skating

Introduction: My name is Kerri Lueilwitz, I am a courageous, gentle, quaint, thankful, outstanding, brave, vast person who loves writing and wants to share my knowledge and understanding with you.