Data is both a critical asset and a potential liability, making it a prime target for threat actors. In fact, the global average cost of a data breach rose to $4.88 million in 2024. With data security now a top priority for organizations worldwide, the demand for advanced DLP (Data Loss Prevention) capabilities has reached an all-time high.
To meet this urgent need, OPSWAT Proactive DLP 2.23.0 introduces powerful new features designed to enhance data protection and keep your organization secure. Leveraging AI-driven technology, this update addresses inappropriate content, expands sensitive data detection, and strengthens data-sharing controls.
Enhanced AI Model for NSFW and Toxic Text Detection
NSFW (Not Safe for Work) images and toxic text - such as hate speech, harassment, or explicit content - are growing risks across industries, from social media platforms to corporate communication channels.
Beyond contributing to hostile work environments and potential legal liabilities, this type of content is also a favored tool among cybercriminals. Attackers exploit it in spear-phishing campaigns, embedding malicious links within inappropriate material to deceive users, or use it in content spoofing attacks that overwhelm moderation systems with harmful content at scale.
Our enhanced AI model addresses these challenges by recognizing subtle cues - such as cultural context and nuanced toxicity - to deliver more effective protection.
Key enhancements include:
- Detects AI-generated NSFW imagery designed to evade traditional classifiers.
- Anonymizes toxic text in files containing violence, political content, drug abuse, or self-harm.
- Supports multiple languages, including English, French, Spanish, German, Italian, and Portuguese.
Expanded AI-Powered PII Detection
Data breaches frequently involve the exposure of PII (Personally Identifiable Information), yet many security tools fail to detect fewer common identifiers that attackers often target. Traditional rule-based detection struggles with complex data patterns, increasing the risk of compliance violations.
Our AI-driven PII detection now supports a significantly expanded range of sensitive data types, making it one of the most comprehensive solutions on the market. This is especially critical in industries such as healthcare and finance, where diverse and high-value PII is both widespread and heavily targeted.
Newly detected data types include:
- Date of birth
- Common medical conditions for PHI (Protected Health Information)
- VIN (Vehicle Identification Number)
- National Health Code Number
- National Drug Code
- Blood Type
- UK Tax ID
- And more
By securing these additional identifiers, organizations can better shield sensitive information, minimize breach risks, and stay compliant with regulations like GDPR, CCPA, and HIPAA.
New Hashing Functionality for PII Substitution
Sharing data for analytics or with third parties requires de-identifying PII to prevent exposure. However, attackers often use MITM (man-in-the-middle) attacks or exploit weak encryption to intercept sensitive information during transmission.
Hashing provides a stronger alternative—as a one-way process with no decryption key, it anonymizes data without compromising its utility.
Proactive DLP 2.23.0 introduces a secure hashing mechanism for PII substitution, enabling safe data sharing while ensuring compliance. Even if intercepted, hashed data remains unintelligible to attackers, adding a critical layer of protection.
Key features include:
- Standardized substitution format:
PII is replaced with<pii-id prefix>_<hash-value> - Supported hashing algorithms:
SHA-256
SHA-384
SHA-512
This functionality enhances security and simplifies adherence to data protection standards, giving organizations confidence in their data-sharing processes.
Addressing Modern Attack Techniques
These new features directly address some of the most pressing cyberthreats organizations face today:
Phishing and Social Engineering
Enhanced NSFW and toxic text detection flags malicious content in emails or messages, reducing the likelihood of users falling for deceptive tactics.
Ransomware and Data Exfiltration
Expanded PII detection identifies and protects a wider array of sensitive data before it’s stolen, while hashing ensures it stays anonymized during sharing.
Deepfakes and Content Manipulation
Our AI-driven detection matches the sophistication of AI-generated threats, catching even the most advanced manipulations.
Reinforced Data Protection with Proactive DLP
As cyberthreats become more frequent and sophisticated, proactive protection is essential. OPSWAT Proactive DLP 2.23.0 empowers organizations to safeguard sensitive data, maintain compliance, and build resilience against modern attack techniques. Whether it's neutralizing inappropriate content, securing diverse forms of PII, or enabling safe data sharing, these enhancements deliver the protection needed to stay ahead of evolving threats.
Additional Resources
Ready to get started with Proactive DLP v2.23.0? Check out these helpful resources:
- Visit opswat.com/technologies/proactive-data-loss-prevention
- Upgrade to Proactive DLP™ v2.23.0
- Learn more about other advanced Proactive DLP features:
- DICOM anonymization
- Not Safe For Work (NSFW) content detection
- Classify PI/PII with OCR and enhanced AI
Have other questions? Reach us at support@opswat.com.
