Liminaの最新情報を読む

Unstructured Data Examples: The Hidden Privacy Risks in Emails, PDFs, and Chat Logs

Most organizations know where their structured data lives. Unstructured data is a different story. Learn where sensitive PII hides in emails, PDFs, chat logs, and clinical notes -- and why pattern matching alone is not enough to protect it.

Manual vs. Automated PII Redaction: Pros, Cons, and When to Use Each

Compare manual and automated PII redaction. Learn the real tradeoffs, compliance implications, and which approach fits your data environment.

Does De-identification Hurt AI Accuracy? The Research Says No

Discover why de-identified data actually enables AI model accuracy. Explore peer-reviewed research on MIMIC, GatorTron, and clinical NLP utility.

Why Health Data Strategies Fail Before They Start

Healthcare data has the potential to transform patient care—but most strategies never get off the ground. Here's why unstructured data, siloed systems, and outdated manual processes are killing your data strategy before it starts, and what to do about it.

Why is linguistics essential when dealing with healthcare data?

Healthcare data is messy, ambiguous, and deeply human. That's why de-identifying it correctly requires more than pattern-matching AI. It requires linguistics. Learn why Limina's linguistics-first approach produces more accurate, context-aware results for healthcare organizations working with sensitive unstructured data.

日本における金融機関の個人情報の取り扱い - 金融庁ガイドラインの要件

The Definitive Guide to Privacy-Preserving Natural Language Processing: Why It Is Essential for the Modern Enterprise

Natural language processing is now central to how enterprises operate, but the data powering these systems is often deeply sensitive. This guide explains why privacy-preserving NLP is no longer optional, how de-identification differs from encryption, and what organizations need to do today to build AI that is secure, compliant, and trusted.

Why The Right To Be Forgotten Is Even Harder To Comply With Than You Think (And What To Do About It)

The right to be forgotten sounds straightforward until you try to actually do it. This article breaks down why erasing a single individual's personal data is technically and operationally complex, and what organizations can do to get ahead of the problem before a deletion request lands in their inbox.

Who is Responsible for Protecting PII?

Protecting personally identifiable information is not the job of any single person or team. This guide breaks down the layered responsibilities of individuals, organizations, and governments, and explains what each must do to prevent data breaches, meet regulatory requirements, and earn the trust of the people whose data they hold.

When is Legitimate Interest a Viable Alternative to Consent?

Consent is not always the only lawful basis for processing personal data. This article breaks down when legitimate interest applies under the GDPR and Canada's proposed Consumer Privacy Protection Act, what the differences are, and the practical advantages and risks for organizations navigating both frameworks.

What the International AI Safety Report 2025 has to say about Privacy Risks from General Purpose AI

The International AI Safety Report 2025, authored by 96 global AI experts, identifies three categories of privacy risk posed by general-purpose AI: risks from training data, risks during active use, and risks from intentional misuse. Here is what the report found, and what it means for organizations handling sensitive data.

What is PHI?

PHI, or Protected Health Information, is a term with a surprisingly wide scope. This article breaks down how HIPAA defines it, how the GDPR and Ontario's PHIPA compare, and why understanding the boundaries of PHI is essential for any organization that handles health data.

What are PII, PCI, and PHI?

In an increasingly digital world where customer data is being collected at various touchpoints, the protection of personal information is becoming increasingly important for businesses worldwide.

What It Really Takes to Build An AI System: It’s more complicated than many think

Building an AI system sounds straightforward until it isn't. From sourcing quality training data to handling real-world edge cases, productionizing models, and maintaining them over time, the true cost of building your own is far greater than most teams anticipate. Here's an honest breakdown of what it actually takes.

WHO’s AI Ethics and Governance Guidance for Large Multi-Modal Models operating in the Health Sector – Data Protection Considerations

The WHO's landmark 2024 guidance on AI ethics for health lays out the privacy risks of large multi-modal models in clinical settings — and charts a path toward responsible deployment. This article breaks down the key risks and recommendations, and what they mean in practice.