Solutions to empower and equip Responsible AI.

Protopia AI’s patented solutions protect the most sensitive information while creating a strong and trustworthy ML lifecycle.

Strengthen the dependability
of your ML lifecycle

Model Training

Protopia AI transforms the training dataset itself. Newly transformed representations of the same data set can be used for training models all without exposing sensitive information.

Model Deployment

Protopia AI creates data transformations specific to your neural network model. Applying these transformations to each data record helps you keep ownership of your data and minimizes the exposure of sensitive information to the model. Your trained model will only see the transformed data and will still operate accurately.

Our solution creates 360° of protection
for your ML tasks.

Protect
the Data

Shrinking the data attack surface keeps raw/plain sensitive information away from humans and ML tasks altogether. By creating transformed versions that can be used without the need to go back to raw/plain form, Stained Glass Transform™ keeps your sensitive information secure.

Protect
the Process

With transformed data being all that ML tasks need to process, this minimizes the risks of model inversion and model misuse overall.

Protect
the Outcome

Stained Glass Transform™ minimizes unintended inferences. Additionally, by transforming inference data, you can enable high quality predictions using sensitive information which would have otherwise potentially not been usable.

Providing protection across
all data types

Protopia AI is the only solution that isn’t tied to any particular ML task or data type. With our first of it’s kind approach, we can perform countless tasks to provide protection. Below you’ll find a few examples of how Protopia works

Stained Glass Transform™ for Inference Data

Choose your data type

Watch Protopia AI in action with the face recognition task below.

Today, the face recognition model would see the data record in its raw form as it appears to the left. This exposes all the information in plain/raw form to the prediction task. With Protopia AI’s Stained Glass Transfrom™ in place, the model will see a completely transformed data record as is shown on the right, but can still make accurate predictions with much less information.

Watch Protopia help an ML task make a prediction on a loan application with the example below.

Traditionally, a model would look at every piece of data in the data record and a combination of all those pieces of information in raw form. This approach enables leaking of sensitive information about the application such as PII. With Protopia in place, the model will read a completely transformed and randomized piece of data but still be able to predict the outcome with the same level of accuracy.

Notice how the personally sensitive attributes on the left are obfuscated on the right with the stochastic embedded representation. In addition, the reconstructed information of text and numbers does not match the actual data depicted on the left. That’s a transformation that automatically occurs in this process.

Watch Protopia in action with an NLP task performing phishing email detection.

However, current NLP tasks have the opportunity to consume data that contains too much sensitive information such as email content. The left side below shows what the tokenized version of emails look like today when using NLP services. On the right, using Protopia’s stochastic transformations, you’ll see how we help protect sensitive information while still enabling an NLP task to identify spam and work accurately.

Visual Data

Watch Protopia AI in action with the face recognition task below.

Today, the face recognition model would see the data record in its raw form as it appears to the left. This exposes all the information in plain/raw form to the prediction task. With Protopia AI’s Stained Glass Transfrom™ in place, the model will see a completely transformed data record as is shown on the right, but can still make accurate predictions with much less information.

Structured/Tabular Data

Watch Protopia help an ML task make a prediction on a loan application with the example below.

Traditionally, a model would look at every piece of data in the data record and a combination of all those pieces of information in raw form. This approach enables leaking of sensitive information about the application such as PII. With Protopia in place, the model will read a completely transformed and randomized piece of data but still be able to predict the outcome with the same level of accuracy.

Notice how the personally sensitive attributes on the left are obfuscated on the right with the stochastic embedded representation. In addition, the reconstructed information of text and numbers does not match the actual data depicted on the left. That’s a transformation that automatically occurs in this process.

Natural Language Processing

Watch Protopia in action with an NLP task performing phishing email detection.

However, current NLP tasks have the opportunity to consume data that contains too much sensitive information such as email content. The left side below shows what the tokenized version of emails look like today when using NLP services. On the right, using Protopia’s stochastic transformations, you’ll see how we help protect sensitive information while still enabling an NLP task to identify spam and work accurately.

Stained Glass Transform™ for Training Data

The Training Data process deal-breaker challenges are rooted in how sensitive training datasets are handled today. Data scientists need access to quality training data to create high performance models that bring real value. Yet they can’t get access to real training data because Data Owners don’t want them to:

  • Query massive plain-text datasets with sensitive information
  • Manually handle and copy datasets with sensitive information
  • Expose sensitive information in training data to arbitrary code written for training.

Stained Glass Transform™ for training data enables

Data Owners to:

  • Enable extracting value from data using ML/AI even when there is entangled sensitive information
  • Give only non-plain-text transformed versions of real data
  • Retain ownership of sensitive information

Data Scientists to:

  • Use real training data to create high performance models that bring real value
  • Maintain data privacy of sensitive information and minimize liability of handling such information
  • Have the ability to use most efficient platform for training without worrying about exposing sensitive information to that platform

Want to see Protopia in action?
Try a demo out today.

23 Comments.