Preventing Unintentional Confidential Information Leaks to Generative AI: Sales Launch of Data Loss Prevention Solution 'Fortra DLP'
Toshiba Digital Engineering has begun selling 'Fortra DLP' starting April 2026. This endpoint security product prevents confidential information leaks caused by prompt inputs and copy-pasting into generative AI, addressing the growing risks of shadow AI and prompt injection.
📋 Article Processing Timeline
- 📰 Published: April 24, 2026 at 18:55
- 🔍 Collected: April 24, 2026 at 10:31
- 🤖 AI Analyzed: April 24, 2026 at 10:57 (26 min after Collected)
■ Confidential information leaks via prompt inputs to generative AI (Note 1) have emerged as a new management risk
■ Control at the "device operation level," such as prompt input and copy-pasting, has become crucial
■ Toshiba Digital Engineering launches sales of "Fortra DLP" as a risk countermeasure assuming the use of generative AI
Toshiba Digital Engineering Corporation (Headquarters: Kawasaki-ku, Kawasaki City, President: Kazuo Mitsuhashi, hereinafter Toshiba Digital Engineering) has started selling "Fortra DLP" from April 2026 to prevent the transmission of confidential information to generative AI prompts.
Background
While the utilization of generative AI contributes to operational efficiency and productivity improvement, the risk of confidential and personal information being unintentionally transmitted externally has become apparent. The "10 Major Information Security Threats 2026" (Note 2) announced by the Information-technology Promotion Agency, Japan (IPA) selected "Cyber risks surrounding the use of AI" for the first time, demanding appropriate AI usage in enterprises. In particular, information leaks due to "Shadow AI," where employees use generative AI at their own discretion, increasingly cannot be prevented by conventional web controls or rule-setting alone, posing a governance challenge.
In addition, the risk of Prompt Injection attacks (Note 3)—embedding malicious instructions into AI to extract confidential information—is also rising, increasing the danger of information leaking externally regardless of employee intent.
In response to these challenges, Toshiba Digital Engineering has launched sales of "Fortra DLP," a DLP (Note 4) product capable of highly accurate visualization and control of data operations on endpoints (PCs).
About "Fortra DLP"
A key feature of "Fortra DLP" is its ability to grasp and control data handling on PCs at the device level, including input operations and copy-pasting to generative AI. Rather than restricting the use of specific generative AI services, it deters the act of confidential data leaving the organization itself, enabling both AI utilization and information protection.
Regardless of the type of browser, application, or cloud service, the following operations can be grasped and controlled at the device level:
■ Input and copy-paste operations to generative AI prompts
■ File copying, moving, and editing
■ Data transfer to USB memory and external storage
■ Email attachments and uploads to cloud storage
In addition to preventing information leaks when using generative AI, "Fortra DLP" can address the following risks, contributing to the strengthening of corporate-wide data governance:
Intentional data removal due to internal fraud
Information leaks due to accidental operations
Data leaks in telework or offline environments
Our company has been handling products from Fortra since 1998, when they operated under the Tripwire brand, and currently has numerous implementation track records as a distributor of Tripwire products.
Leveraging the technical expertise and implementation know-how cultivated over many years, we provide consistent support from implementation consideration to design, deployment, and operation of "Fortra DLP."
Furthermore, Toshiba Digital Engineering offers the following support menus:
■ Fortra DLP Cloud Implementation Support Service
■ Fortra DLP On-premises Server Construction Service
■ Fortra DLP Administrator Training Service
■ Fortra DLP Maintenance Support Service
With these unique services, even companies that cannot secure sufficient specialized security or IT personnel can smoothly proceed from implementation to operational entrenchment.
Note 1: Information leak caused by prompts containing confidential or personal information being transmitted to external generative AI services
Note 2: 10 Major Information Security Threats 2026 [Organizations]
Rank | Threat for "Organizations" | First Selected Year | Mentions in Top 10 (Since 2016)
1 | Damage caused by ransomware attacks | 2016 | 11th consecutive year
2 | Attacks targeting supply chains and contractors | 2019 | 8th consecutive year
3 | Cyber risks surrounding the use of AI | 2026 | First selection
(Source: Excerpt from "10 Major Information Security Threats 2026" by the Information-technology Promotion Agency, Japan (IPA))
Note 3: A method of extracting information by giving malicious instructions to AI, causing unexpected behavior (Prompt Injection)
Note 4: Data Loss Prevention
Information leak by Prompt Injection attack
Prevention of information leak by "Fortra DLP"
Details on "Fortra DLP"
This service...
■ Control at the "device operation level," such as prompt input and copy-pasting, has become crucial
■ Toshiba Digital Engineering launches sales of "Fortra DLP" as a risk countermeasure assuming the use of generative AI
Toshiba Digital Engineering Corporation (Headquarters: Kawasaki-ku, Kawasaki City, President: Kazuo Mitsuhashi, hereinafter Toshiba Digital Engineering) has started selling "Fortra DLP" from April 2026 to prevent the transmission of confidential information to generative AI prompts.
Background
While the utilization of generative AI contributes to operational efficiency and productivity improvement, the risk of confidential and personal information being unintentionally transmitted externally has become apparent. The "10 Major Information Security Threats 2026" (Note 2) announced by the Information-technology Promotion Agency, Japan (IPA) selected "Cyber risks surrounding the use of AI" for the first time, demanding appropriate AI usage in enterprises. In particular, information leaks due to "Shadow AI," where employees use generative AI at their own discretion, increasingly cannot be prevented by conventional web controls or rule-setting alone, posing a governance challenge.
In addition, the risk of Prompt Injection attacks (Note 3)—embedding malicious instructions into AI to extract confidential information—is also rising, increasing the danger of information leaking externally regardless of employee intent.
In response to these challenges, Toshiba Digital Engineering has launched sales of "Fortra DLP," a DLP (Note 4) product capable of highly accurate visualization and control of data operations on endpoints (PCs).
About "Fortra DLP"
A key feature of "Fortra DLP" is its ability to grasp and control data handling on PCs at the device level, including input operations and copy-pasting to generative AI. Rather than restricting the use of specific generative AI services, it deters the act of confidential data leaving the organization itself, enabling both AI utilization and information protection.
Regardless of the type of browser, application, or cloud service, the following operations can be grasped and controlled at the device level:
■ Input and copy-paste operations to generative AI prompts
■ File copying, moving, and editing
■ Data transfer to USB memory and external storage
■ Email attachments and uploads to cloud storage
In addition to preventing information leaks when using generative AI, "Fortra DLP" can address the following risks, contributing to the strengthening of corporate-wide data governance:
Intentional data removal due to internal fraud
Information leaks due to accidental operations
Data leaks in telework or offline environments
Our company has been handling products from Fortra since 1998, when they operated under the Tripwire brand, and currently has numerous implementation track records as a distributor of Tripwire products.
Leveraging the technical expertise and implementation know-how cultivated over many years, we provide consistent support from implementation consideration to design, deployment, and operation of "Fortra DLP."
Furthermore, Toshiba Digital Engineering offers the following support menus:
■ Fortra DLP Cloud Implementation Support Service
■ Fortra DLP On-premises Server Construction Service
■ Fortra DLP Administrator Training Service
■ Fortra DLP Maintenance Support Service
With these unique services, even companies that cannot secure sufficient specialized security or IT personnel can smoothly proceed from implementation to operational entrenchment.
Note 1: Information leak caused by prompts containing confidential or personal information being transmitted to external generative AI services
Note 2: 10 Major Information Security Threats 2026 [Organizations]
Rank | Threat for "Organizations" | First Selected Year | Mentions in Top 10 (Since 2016)
1 | Damage caused by ransomware attacks | 2016 | 11th consecutive year
2 | Attacks targeting supply chains and contractors | 2019 | 8th consecutive year
3 | Cyber risks surrounding the use of AI | 2026 | First selection
(Source: Excerpt from "10 Major Information Security Threats 2026" by the Information-technology Promotion Agency, Japan (IPA))
Note 3: A method of extracting information by giving malicious instructions to AI, causing unexpected behavior (Prompt Injection)
Note 4: Data Loss Prevention
Information leak by Prompt Injection attack
Prevention of information leak by "Fortra DLP"
Details on "Fortra DLP"
This service...