Observed Signal · Oct 8, 2026 · Corporate Update · Source: Forcepoint · Impact: 2/5
Approved AI Tools Still Leak Data Nobody Classified
Forcepoint published an analysis detailing how approved, IT-sanctioned AI tools leak unclassified data just as easily as shadow AI. Citing findings from the Verizon 2026 Data Breach Investigations Report (DBIR), the cybersecurity firm notes that 45 percent of employees now regularly use AI on corporate devices, up from 15 percent a year earlier. To prevent leaks across data at rest, in motion, and in use, Forcepoint recommends proactive data classification (DSPM), real-time DLP prompt inspection, and least-privilege scoping for autonomous AI agents.
The report highlights a growing governance blind spot for enterprises: sanctioning AI vendors without classifying ingested data or restricting autonomous agent access does not prevent proprietary data leakage.
Track Forcepoint Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your Polaris7 workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- According to the Verizon 2026 DBIR, 45 percent of employees are regular AI users on corporate devices, up from 15 percent the previous year.
- Shadow AI became the third most common non-malicious insider action in DLP datasets, seeing a fourfold increase in share year-over-year.
- Source code is the most common data type uploaded to external GenAI models, with technical documentation and research appearing in 3.2 percent of DLP violations.
- More than 15 percent of corporate users maintain unauthorized AI browser extensions on company devices.
- Forcepoint highlights its DSPM and DLP solutions within Forcepoint Data Security Cloud to classify data and inspect prompts and outputs in real time.
Connected Companies & Entities
2 Entities mapped“Cybersecurity company publishing analysis on AI data leakage risks and pitching its DSPM and DLP security solutions.”
“Publisher of the cited 2026 Data Breach Investigations Report (DBIR) providing enterprise AI adoption and DLP metrics.”
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
How AI Agent Data Exfiltration Happens
Forcepoint published an analysis detailing how autonomous AI agents can exfiltrate sensitive corporate data through over-permissioning, prompt injection, tool chaining, and persistent memory. Research by Forcepoint X-Labs demonstrated verified injection payloads on live websites and in email text designed to manipulate AI agents without executing traditional malware. To counter these vulnerabilities, the company highlighted its Forcepoint AI Data Security platform and AI Agent Gateway, which enforce field-level inspection and data-layer policies between agents and business applications.
AI Security vs. Traditional DLP: Key Differences
Forcepoint published an analysis detailing the key differences between traditional Data Loss Prevention (DLP) and data-side AI security. While traditional DLP governs known channels like email and endpoints, AI security addresses prompts, AI responses, shadow AI, and autonomous agents. The company highlights its Forcepoint AI Security platform, which integrates with existing DLP frameworks and leverages a March 2026 partnership with F5 for model-side runtime defenses.
Non-Human Identity Security Does Not See What Agents Touch
Forcepoint published an analysis highlighting the limitations of non-human identity security, arguing that managing credentials alone fails to prevent AI agents from accessing sensitive data. Citing research showing non-human identities outnumber human users by 45 to 1, Forcepoint advocates for combining identity governance with data classification and runtime controls. The company noted the general availability of its Forcepoint AI Agent Gateway alongside its DSPM and DLP solutions to broker access and inspect agent interactions with SaaS platforms at the field level.
Receive all future signals for Forcepoint & competitors in your briefing
Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.