Urgent.News

What's breaking now, across hundreds of outlets.

AI

Shadow AI in the Workplace: Data Security Risks and Governance Strategies

Workplace AI bans may not stop Shadow AI. Explore the risks to confidential data and practical ways companies can govern employee AI use.

Shadow AI in the Workplace: Data Security Risks and Governance Strategies

It is early hours of the morning. A junior associate is trying to finish a lengthy agreement by 8:00 AM. The coffee is cold. The associate uses a free AI chatbot on their personal phone to quickly summarize three clauses from the document. This makes the work easier, but the firm has banned the use of public AI tools. The associate has found a workaround by accessing the tool through their personal device.

Shadow AI refers to employees using AI tools without organizational knowledge or approval. These tools can include chatbots, writing assistants, browser extensions, coding tools, and meeting transcription applications. The issue is not just the tool itself, but what information the employee shares with it. Even seemingly harmless actions like uploading client documents for summaries or entering company data for reports pose risks. The organization may have no idea where the data goes or how it is stored.

Many companies have responded by blocking public AI platforms. They are worried about confidential information, customer records, and business documents being shared with unassessed services. However, blocking these platforms does not necessarily stop employees from using them. If a company bans an AI website but expects employees to complete their work on time, the need for AI may persist. Some may simply find alternative ways to access it, such as through personal devices.

Employees use AI tools for practical reasons like saving time or efficiently completing tasks. However, this does not make all uses acceptable. Employees still need to follow policies and protect confidential information. Organizations need to understand why employees turn to unapproved tools and address their concerns about which tools are acceptable, what information can be shared, and how to handle tasks when approved systems are insufficient. Without clear answers, employees may make their own decisions, leading to potential data risks.

Rewritten by urgent.news; may contain errors — check the original.

Source: hackernoon.com

Read the original at hackernoon.com →

Report a problem with this story

More in AI

More from Saturday 14 November →