Oversharing of sensitive documents soars 450%

There was 450% increase in the oversharing of sensitive documents over past year, according to recently released research from Concentric.

The AI-based solutions vendor released its Q1 2021 quarterly Data Risk Report, which highlights a continued rise in oversharing of business-critical and sensitive documents.

The report revealed the number of overshared files rose 450% compared to the same quarter in 2020, highlighting the significant impact of the pandemic and remote work on data security.

Using its Semantic Intelligence solution, Concentric captured user data in production deployments from companies in the technology, financial, and healthcare sectors to reveal how organizations create, use, and manage data.

The company scanned more than 110 million unstructured data files to discover business-critical and sensitive documents that are overshared via link sharing, inappropriate external sharing, internal permission misconfigurations, and incomplete/incorrect document classifications.

Oversharing increases the risk an organisation will lose data, violate compliance or privacy mandates, or experience cybercrime. Statistics highlighted in this latest Data Risk Report reveal that organisations average 439,000 files at-risk due to oversharing. That translates to 210 at-risk files per employee (up significantly from 38 files per employee in Q1 2020, a 452%increase).

Link-based risky sharing is up to 65,000 documents per enterprise, up from 56,000 in Q4 2020 (the company started tracking link sharing risks in Q3 2020).

Concentric is the first company to identify and quantify risk in both structured and unstructured data using deep learning. Its solution autonomously provides an accurate and detailed semantic understanding of the millions of contracts, financial documents, payroll, M-A plans, product roadmaps and source code files used by organisations every day.

Similar to previous Data Risk Reports, this new report analysed production data and reflects actual user practices and real-world data risk exposures.

Additional findings include:

• Nearly 35% of unstructured data is business-critical - that's 3.1 million files in an average organisation. Of those business-critical files, 14% can be seen by internal or external users who should not have access. 
• 229,000 business-critical files were erroneously classified and inappropriately accessible by other employees. To illustrate, nearly 23% of all unstructured data contained PII and were not marked appropriately. 
• More than 33% of files processed were duplicates (15%) or near-duplicates (20%). Maintaining multiple variant copies of sensitive information (often with insecure file permissions, prohibited locations, or improper file classifications) can create legal and regulatory risks, as well as significant unnecessary storage costs. 
• Eighty five percent of at-risk files were overshared with users or groups within the company, while 15% of business-critical files were overshared with external third parties. 

"More than 80% of an organisation's data is unstructured, meaning it's embedded in the millions of financial reports, corporate strategy documents, source code files, and contracts created every year," says Karthik Krishnan, chief executive at Concentric.

"As our quarterly Data Risk Reports continue to show, unstructured data is still largely unseen, unexplored and insecure, and too often overshared inside and outside organisations," he says.

"This educational report for the industry demonstrates the power of Concentric's advanced deep-learning technology to solve the ongoing challenges of unstructured and structured data security - both in the cloud and on-premises.

"Our Semantic Intelligence solution autonomously delivers the data risk insights enterprises need to help employees work productively - remotely or in the office - without sacrificing security."

Concentric's Risk Distance analysis evaluates business criticality based on contextualised content, file ownership, document meta-data, presence of personally identifiable information (PII), and peer file comparisons.

Business criticality is vital to security assessment and understanding which files must not be overshared. Product files accounted for the leading share of business-critical documents (44%) analysed for this report, followed by financial files (27%), legal files (13%), and partner documents (10%).