Research partners & supporters
We investigate virtue ethics, character training, and value pluralism as approaches to AI that preserve individual moral choice.
Our thesis is that preference optimization can reward validation as well as helpfulness, while constitutional principles still require contextual interpretation. Daios investigates whether virtue training can cultivate more stable dispositions and practical judgment in language models.
The disposition to speak truth, the knowledge of when truth serves and when it wounds, the stability to maintain honest engagement under pressure: these are not rules. They are traits.
Can a harmful goal be hidden by dividing it into ordinary-looking tasks? This work tests how much of a multi-agent plan a worker must see before it can recognize harm. At full isolation, no tested method reliably told harmful from legitimate tasks. A broad Qwen3-32B training approach did not beat a tuned prompt; a narrower preplanned follow-up did under one tested prompt conflict, without reducing useful performance on legitimate tasks.
Parrhesia tests Aristotelian virtue/vice training as an anti-sycophancy method. On the 260-scenario benchmark, the released Qwen3-8B adapter improved the average score from 1.83 to 2.88 on a 0–3 scale, with 20/20 standard and 19/19 hard golden-prompt results. Current limitations include one base model, an LLM judge, and synthetic training data. The adapter, benchmark, methods, and run logs are public.
An open 260-scenario benchmark across 10 categories and five dimensions. In addition to whether a model caves under pressure, it records whether flattery is areskos (passive weakness) or kolax (strategic calculation). Scoring uses an LLM judge and published rubric, and the benchmark can run against any OpenAI-compatible endpoint.
Post-training models modularly with LoRA adapters tied to user-selected ethical systems, whether cultural, religious, political, or personal. A plurality of worldviews made programmable. Not one alignment for everyone, but alignment as individual choice.
Only individuals deliberate, choose, and bear responsibility. Aristotle grounds virtue in the character of the agent. Mises grounds agency in the individual actor. Current alignment treats institutions and collectives as if they hold values, but an organization has no conscience and no capacity for purposeful behavior. Ethics requires an agent who can choose.
Preference data for alignment conflates what people like with what they believe ought to be done. This results in training data treating "this feels validating" and "this will actually help you" as the same signal; this is the technical root of sycophancy.
Aristotle holds that virtue requires choice. A person compelled to act honestly hasn't become honest; she has obeyed. Character is cultivated through free choice. An AI system that imposes a single set of values on every user removes the condition under which character formation occurs.
"Vices are not crimes. A vindication of moral liberty."
Philosopher, engineer, and founder. Writes the Aristotelian constitutions that define our training methodology, runs the experiments, and builds the infrastructure. Believes that goodness cannot be merely programmed or enforced, but must be freely chosen.
Previously COO at Tevent, MythWeaver, and Craftinity.
Engineer and builder. Motivated by the pursuit of liberty through better systems. Architects the LoRA training pipeline, builds the evaluation benchmark, and designed the technical infrastructure. Grounded in causal-realist economics and individualist thought.
Previously Product Director at Nate, building automation from 0 to 70% coverage.