About Me
I am a governance architect and AI strategist committed to guiding organizations through the responsible integration of AI technologies, not through hype, but through operational clarity, lifecycle governance, and measurable trust.
I specialize in diagnosing the systemic fragilities of large language models, mapping ethical risks across the AI data lifecycle, and designing scalable frameworks to counter bias, misinformation, and error propagation. My research, including the MIDCOT Framework, BME Metric Suite, and SymPrompt+, bridges standards-based governance with hands-on deployment, enabling organizations to align AI tools with real-world intent and regulatory expectations.
​
In a world flooded with AI misinformation and false expertise, I aim to be a rational, standards-aligned voice, demystifying agentic AI, promoting ethical augmentation, and anchoring innovation in accountability.
My research explores how Large Language Models (LLMs) amplify bias, misinformation, and error (BME) through feedback loops, model drift, and human-in-the-loop cognitive echo chambers. These effects, while often subtle, can degrade the trustworthiness of AI systems and compromise decision-making, governance, and data quality at scale.
To counter these risks, I developed the BME Metric Suite, which includes:
-
Bias Amplification Rate (BAR)
-
Echo Chamber Propagation Index (ECPI)
-
Information Quality Decay (IQD)
-
Plus, emerging metrics such as PTDI and AHRS
These metrics provide a quantifiable lens for detecting, assessing, and governing the health of AI systems over time.
I also designed the SymPrompt Framework, a structured approach to ethical prompt engineering. It reduces bias drift, enhances alignment, and fosters transparency between human users and generative models, forming the foundation for a new paradigm of symbiotic prompting.
My work is anchored in internationally recognized standards, including:
-
ISO/IEC 42001 (AI Management)
-
ISO 8000 (Information Quality)
-
ISO/IEC 27001/27701 (Security & Privacy)
-
ISO 23053
-
NIST AI Risk Management Framework
​
My goal is simple but urgent: to equip organizations with practical, rigorous tools for navigating the ethical integration and lifecycle governance of AI systems, before the cost of inaction becomes irreversible.