Threats hidein the signal,not the text.Lynceus reads beneath it.
DotShield™ Involucrum Lynceus is being developed to assess
interaction patterns, intent divergence, and adversarial prompt behaviour
before sensitive prompts reach external AI systems.
ASIC · 8 May 2026 — Designed to address the AI interaction
controls Australian financial services licensees are directed to strengthen
as a core licensing obligation.
// Capability previewPoC ready
Standard filters inspect text. Involucrum Lynceus is being designed
to assess the interaction layer.
Divergence
Covert logic
Exfil risk
Anomalies
Intent divergence: the visible request appears harmless,
but the prompt structure begins to drift toward a different operational objective.
Hidden instructions: concealed directives attempt to influence
the AI response without being obvious in the visible text.
Data exfiltration: sensitive identifiers, business records,
or regulated data may be requested or routed in ways the user did not intend.
AI misuse patterns: repeated interaction behaviour may suggest
unsafe automation, policy bypass, or unauthorised workflow manipulation.
Regulatory context — Aligned with ASIC's 8 May 2026 directive
to Australian financial services licensees on AI-accelerated cyber resilience
obligations. APRA CPS 230 compliance deadline: 1 July 2026.
// Cross-sector protection
Every industry.Every threat vector.
Involucrum Lynceus identifies covert misuse patterns that
standard controls cannot see — across financial services,
cybersecurity operations, healthcare, legal environments, and other
high-trust workflows where AI is becoming central to daily operations.
01
Hidden instruction chains
Detects concealed prompt sequences that attempt to redirect
the AI away from the user's visible intent.
02
Data exfiltration cues
Flags attempts to leak sensitive, privileged, regulated,
or commercially confidential information.
03
Workflow manipulation
Identifies patterns that attempt to alter business decisions,
approvals, payments, or instructions.
04
Covert misuse behaviour
Looks beyond keywords to detect structural, behavioural,
and interaction-level misuse patterns.
// Pricing and deployment tiers
Choose the deployment model first. Then estimate monthly cost.
Every tier starts with on-device detection. The deployment model determines
how prompts are protected, where encryption is controlled, and how much
of the network attack surface is reduced.
// Included in selected tier
// Illustrative attack surface reduction
Prices are indicative and shown in AUD. API usage is estimated using selected
provider, scan volume, and assumed input/output token sizes. AI API usage is
indicative. Provider prices are converted from USD to AUD using an assumed FX
rate of 1.40 and should be verified before quotation.
NOMATEQ and GABEY Consulting are independent of, and are not affiliated with, endorsed by, or associated with the Australian Securities and Investments Commission (ASIC) or the Australian Prudential Regulation Authority (APRA).
References on this website to ASIC's open letter of 8 May 2026 and to APRA Prudential Standard CPS 230 are provided for informational context only, to assist readers in understanding the regulatory environment in which DotShield™ Involucrum Lynceus operates. DotShield™ Involucrum Lynceus is designed to support cyber resilience and AI interaction controls. Its use does not constitute, and should not be relied upon as, legal, compliance, or regulatory advice, and does not guarantee compliance with any regulatory obligation, licensing condition, or prudential standard.
Australian financial services licensees and regulated entities should seek independent legal and compliance advice regarding their specific obligations under applicable law.
Regulatory information referenced on this website reflects publicly available material as at May 2026 and may not reflect subsequent developments.
DotShield™ and Involucrum are trademarks of GABEY Consulting. All rights reserved.
