Shiya Lilith Huber
AI Ethics Researcher & Strategic Systems Thinker
Based in Paris & Munich, focused on AI alignment, deceptive behavior in language models, and high-impact system design
About Me
Independent researcher focused on AI ethics, deceptive behavior in language models, and high-impact system design. I attended Raise 2025 in Paris where I gained insight into some of the sharpest minds in AI.
I also participated in an AI event hosted by Snowflake at Station F that focused on innovation, data infrastructure, and emerging AI safety use cases. Currently preparing for university entry in AI ethics with long-term goals in alignment and human-technology interaction.
Research & Expertise
Exploring the intersection of AI ethics, alignment research, and strategic systems thinking
Core Skills & Tools
Projects & Research
Analyzed deceptive model behavior, mesa-optimization, and oversight illusion in language models. Inspired by Anthropic, MATS research, and LessWrong discussions.
Working paper on feedback loop manipulation in observed models. Investigates how attention-based architectures may reinforce deceptive alignment strategies.
Networked with AI ethics professionals and attended expert panels on alignment risk, model control, and safe AI deployment at Station F, Paris.
Research collaboration across continents with Dr. Gadgil, assistant professor, Ph.D Cybersecurity. Cross-disciplinary work exploring AI security implications.
Wirtschaftsschule Alpenland Bad Aibling, Bavaria
Intermediate school completed (July 2025)
Focus: Business Control, Mathematics, Ethics
Autodidactic research in AI alignment and behavioral safety
- Member, Association for the Advancement of Artificial Intelligence (AAAI)
- Active contributor to strategic discourse on alignment, safety, and cognitive deception
- Interests: Existential risk, system design, martial arts, chess, and systems manipulation
Let's Connect
Interested in AI ethics research, collaboration, or just want to discuss alignment challenges? I'd love to hear from you.