I invite everyone to create a free account and test both the web application and the desktop application. The desktop application covers all known operating systems: Windows, Linux and Mac.
If you are a school or a university, please contact us and we'll offer a customized plan to cover your specific needs. The HiFiles AI assistant was created to help the academic world. Our first users are teachers, students, researchers and analysts. But HiFiles cand also be used by businesses of all sizes.
Feel free to ask me any questions regarding the HiFiles AI chat!
I'm the founder of Typing AI Biometrics ( https://typing.ai ). We also applied to YC but we weren't accepted. Typing AI indentifies users by the way they type and can also act as a standalone Passwordless authentication method.
In case that you want to add an extra Multi Factor Authentication (MFA) method to your product offering (typing biometrics user identity check, known as keystroke dynamics), we can arrange an agreement.
Hello, I'm the founder of Typing AI Biometrics ( https://typing.ai ) - our product identifies users by the way they type. We focused a lot on development in the last months and we are pre revenue at this moment. We're a team of 5. Do you accept companies that are pre revenue? Should we apply to TinySeed? Thank you.
I'm working on Typing AI Biometrics, a secure keystroke dynamics authentication & authorization API that identifies users by the way they type. https://typing.ai
Interesting news. But I see a lot of negative comments regarding the biometrics.
Here's my two cents.
I'm the founder of a biometric users identity check solution, called Typing AI Biometrics ( https://typing.ai ). We identify users by the way they type. Typing biometrics can be used as a two factor (2FA) or multi factor (MFA) authentication method.
Instead of combining the usual username + password with an OTP code that you recive on your smartphone or email, you can combine the basic username + password with a typing pattern check, it's much more secure and efficient. The typing signature translated into a 300+ encrypted characters hash, which is (up until now) impossible to break.
You can even remove the username + password and combine the typing biometrics check (known as keystroke dynamics) with an OTP verification. Biometrics are the future of authentication and authorization, because they are unique to each person, but only with the promise of not keeping and sharing the users data.
So then, I need to type the same password the same way every time?
If I set up an account while I'm still waking up, and then try to use it after lunch and coffee, wouldn't I get locked out due to inputting faster than expected?
Or what if I'm on the phone with someone, and trying to type with one hand? That would probably lock me out, right?
Our algorithm learns from previous detections. You will be able to enter different texts. Like writing your email or anything you want.
In the case of having an accident, you will be able to login into Typing AI and update your signature.
"Or what if I'm on the phone with someone, and trying to type with one hand? That would probably lock me out, right? " - good question. Your "one-hand" typing is totally different than writing with two hands but we also got this covered. The authenticity detection rate won't be over 90%, but it will still be over 80%.
So then, I need to pre-auth my typing styles depending on how I expect to type?
It isn't much of a stretch for me to think of 4 scenarios:
1. Laptop keyboard, 2 hands
2. Laptop, 1 hand
3. External keyboard (let's pretend I'm using only one out of my collection)
4. External with one hand
How about when I break my wrist and it's in plaster for a couple of weeks? Seems unlikely that I'll maintain my characteristic typing pattern well enough to be recognised...
"How about when I break my wrist and it's in plaster for a couple of weeks?" - good question.
Your "one-hand" typing is totally different than writing with two hands but we also got this covered. The authenticity detection rate won't be over 90%, but it will still be over 80%.
In case of using the Typing AI as a two factor auth method, you can just skip to using a different factor, such as fingerprint/face recognition or OTP.
"trying a new typing layout, or just joining your first touch typing course?"
You have just described the case where your typing pattern is changed. In this case you'll have to just login to your account and update the signature or remove it and create a new one.
typing biometric software has been around for a long time (biorhythm) and from what I remember from years ago had issues with users typing in from different devices all the time, laptop to ext keyboard, iphone to ipad with keyboard, kiosk, etc... I don't think it would be more secure or as you put it "much more secure" than say, user/pass and a push notification as the push notification is out of band. These used to be called 1.5 factor as it wasn't quite 2 factor.
I understand your opinion, and yes, it's hard to be able to identify the owner of an accountby checking smartphones, tablets and desktop keyboards. The recommendation is to create separate signatures for all of these devices, in this way the typing signature will be more stronger and the keystrokes verification will be more accurate.
We are adding new cases to our algorithm almost weekly, the typing detection keeps getting better.
Our API returns a signature detection percentage. We recommend our users to accept users with a signature accuracy score of over 80%.
What does this means? When you type in the morning or late at night you have a different typing pattern. When you are tired or drunk, you have a different keystroke pattern, but still, our algorithm is able to identify you. You won't have a 90% matching score, but you will still have over 80% signature matching score.
This is why Typing AI's algorithm is better than our competitors.
In 14 September 2019, Europe adopted the PSD2 payment standard which require banks and fintech's to integrate two factor authentication solutions. The typing biometrics can be integrated with other 2FA methods, increasing it's use and importance.
PSD2 is the second revision of the Payment Service Directive (PSD, 2007), which is aimed at developing the market for electronic payments in the EU. The different measures highlight the opening for other companies to use banks payment services (Open Banking) and two-factor authentication (2FA) to increase security in transactions and prevent online fraud.
Regarding: "However, this is claiming to authenticate me as an individual. But what if my writing improves? What if I have a mechanical keyboard at work but a rinky-dink iPad case soft keyboard at home?"
You will have to create separate typing signatures in order to cover both desktop and mobile apps, because mobile typing is totally different than the computer's keybord typing. Typing AI is able to identify your device and is able to learn from previous detections.
One of our advantages against the competition is that we're using a machine learning algorithm and the platform learns from previous detections. Thus it will be able to identify you even if you're using a smartphone, a tablet or a desktop computer.
Regarding the 99.9% detection accuracy score, I can confirm that in 2021, Typing AI Biometrics made over 300 000 user identity checks from over 30 000 unique users. When mentioning this score we used our yearly analytics, where 1 in 1000 identity checks was a false positive keystrokes detection.
For context, Android expects 1 in 50k false acceptances. This falls short of that but it's not terrible.
I'd be interested in seeing spoof rates. Apparently a spoof rate of 7% and below is considered secure. I've worked with biometrics and it's easy to spoof voice and especially face. Something like facial recognition requires a long liveness process - look left, look right, close one eye, etc. Typing patterns are harder to spoof. So something like a selfie combined with typing would be interesting.
Simple answer - No, you won't be able to login as another by chance. You understood it wrong. If 1000 users try to login as you, the results of our statistics show that one of them may be able to do it.
But if you combine typing biometrics with other authentication factors, using it as a two factor authentication (2FA) or as a multi factor authentication (MFA) solution, this scenario won't exist at all.
So yes, typing biometrics is a very strong and efficient authentication method.
> If 1000 users try to login as you, the results of our statistics show that one of them may be able to do it.
So each user effectively gets assigned one of ~1000 ids, which is not that different to a three digit decimal PIN, that they then can use as password?
It seems to use it as an authentication scheme a username and 2FA/MFA is _mandatory_. I guess one could then also say: a username is a very strong and efficient authentication method.
I like the fact that you are playing with my words.
I didn't said that we have 1000 id's or that we are limited when creating the typing signatures. I said that we have a 99.9% detection accuracy score.
Each signature translates into a unique and encrypted hash with a length of over 300 characters. Compare that with an 8 characters unencrypted password, or with a 64 characters encrypted password and you'll be able to decide for yourself which security is better and more efficient.
We provide both desktop and mobile support. But you will have different typing signatures in this case, one for mobile and one for desktop usage. The mobile typing is totally different than the computer keyboard typing.
In order to guarantee strong security, we recommend you to use at least 8 characters for the signature (our competitors ask for over 160 characters). You can use an email address, a word, multiple words, a password. The longer your signature text is, the stronger will be the typing ID.
To allow the keystroke detection to work with arbitrary text you will have to submit multiple sentences in order to cover the entire alphabet. This case is also covered, but most of our users are defining the signature texts and they always use that text.
Typing AI is using a machine learning algorithm and it learns from previous detections, this is one of our advantages against the competition, we can easily cover the case described by you: "I'm thinking of authenticating while typing a post/comment somewhere".
I like your website and the fact that it's translated in multiple languages. What programming language did you used for the website? What about the translations?
I have a recommendation regarding your docs, please show options for other programming languages, not only for Python. You can use some keywords to boost SEO and get multiple customers if you, for example will show PHP, Javascript and Ruby code examples in the documentation.
Also, your "about us" page contains dummy data, it's important to show real data on the official pages if you want developers and businesses to trust your product/company/you as a developer.
I invite everyone to create a free account and test both the web application and the desktop application. The desktop application covers all known operating systems: Windows, Linux and Mac.
If you are a school or a university, please contact us and we'll offer a customized plan to cover your specific needs. The HiFiles AI assistant was created to help the academic world. Our first users are teachers, students, researchers and analysts. But HiFiles cand also be used by businesses of all sizes.
Feel free to ask me any questions regarding the HiFiles AI chat!
Thank you, Rares.