Independent Review Certifies Private AI’s PII Identification Model as Secure and Reliable

Share This Post

An independent evaluation has confirmed that Private AI’s Personal Identifying Information (PII) identification model is both highly effective and secure, now backed by Armilla AI’s Guaranteed Warranty and insurers Swiss Re, Greenlight Re and Chaucer. This validation addresses crucial concerns regarding the reliability and fairness of AI-driven privacy tools, providing assurance to organizations that their data privacy solutions meet the highest standards.

Private AI’s PII identification model has now received the Armilla AI Guaranteed Warranty, reflecting the comprehensive assessment which highlights:

  • • Performance Baseline: Baseline performance: Private AI’s PII identification model misses less than 1% of PII as a percentage of total word on a comprehensive test set of 109,000 words consisting of text extracts from different sources such as financial transactions, medical transcripts, Twitter posts, internet forum posts, miscellaneous texts, and correspondence.
  • • Bias and Fairness: The independent assessment found no significant biases in the model’s performance across various demographic names. This ensures that the identification process is equitable and unbiased, maintaining fairness across all datasets.
  • • Robustness: The model was rigorously tested against real-world challenges, including OCR errors and text corruption. It maintained reliable performance, demonstrating resilience in handling data quality issues.
  • • Organizational Considerations: Private AI has shown a commitment to high standards of data governance and responsible AI deployment, with thorough testing and validation protocols in place.

“Receiving the Armilla AI Guaranteed Warranty is a significant milestone for us,” said Patricia Thaine, CEO & Co-Founder of Private AI. “It not only reinforces our commitment to delivering a top-tier data privacy solution but also provides our clients with a strong assurance of our product’s reliability and effectiveness.”

PII redaction is crucial in today’s regulatory environment, where privacy and compliance are top priorities. The independent validation of Private AI’s model ensures that organizations can trust their privacy tools to meet stringent performance and fairness standards.

The assessment involved rigorous testing on various datasets, including those from the medical domain and structured data formats. Metrics such as precision, recall, and missed PII percentage were evaluated, along with the model’s adaptability to real-world data conditions.

The conclusion of the assessment confirms that Private AI’s PII identification model is a robust solution for data privacy. With the added security of Armilla AI’s Guaranteed Warranty, clients can be confident in the model’s performance and reliability.

For organizations looking to enhance their data privacy measures, Private AI’s model, now guaranteed by Armilla AI, represents a dependable and effective solution.

To learn more about how Private AI’s technology can protect your data, contact us today.

Subscribe To Our Newsletter

Sign up for Private AI’s mailing list to stay up to date with more fresh content, upcoming events, company news, and more! 

More To Explore

Download the Free Report

Request an API Key

Fill out the form below and we’ll send you a free API key for 500 calls (approx. 50k words). No commitment, no credit card required!

Language Packs

Expand the categories below to see which languages are included within each language pack.
Note: English capabilities are automatically included within the Enterprise pricing tier. 

French
Spanish
Portuguese

Arabic
Hebrew
Persian (Farsi)
Swahili

French
German
Italian
Portuguese
Russian
Spanish
Ukrainian
Belarusian
Bulgarian
Catalan
Croatian
Czech
Danish
Dutch
Estonian
Finnish
Greek
Hungarian
Icelandic
Latvian
Lithuanian
Luxembourgish
Polish
Romanian
Slovak
Slovenian
Swedish
Turkish

Hindi
Korean
Tagalog
Bengali
Burmese
Indonesian
Khmer
Japanese
Malay
Moldovan
Norwegian (Bokmål)
Punjabi
Tamil
Thai
Vietnamese
Mandarin (simplified)

Arabic
Belarusian
Bengali
Bulgarian
Burmese
Catalan
Croatian
Czech
Danish
Dutch
Estonian
Finnish
French
German
Greek
Hebrew
Hindi
Hungarian
Icelandic
Indonesian
Italian
Japanese
Khmer
Korean
Latvian
Lithuanian
Luxembourgish
Malay
Mandarin (simplified)
Moldovan
Norwegian (Bokmål)
Persian (Farsi)
Polish
Portuguese
Punjabi
Romanian
Russian
Slovak
Slovenian
Spanish
Swahili
Swedish
Tagalog
Tamil
Thai
Turkish
Ukrainian
Vietnamese

Rappel

Testé sur un ensemble de données composé de données conversationnelles désordonnées contenant des informations de santé sensibles. Téléchargez notre livre blanc pour plus de détails, ainsi que nos performances en termes d’exactitude et de score F1, ou contactez-nous pour obtenir une copie du code d’évaluation.

99.5%+ Accuracy

Number quoted is the number of PII words missed as a fraction of total number of words. Computed on a 268 thousand word internal test dataset, comprising data from over 50 different sources, including web scrapes, emails and ASR transcripts.

Please contact us for a copy of the code used to compute these metrics, try it yourself here, or download our whitepaper.