Become an AI engineer

# Truthfulness

Truthfulness is a unit of 4 lessons in Rabbithole's Become an AI engineer course, at levels 3 to 4.

## Lessons

1. Hallucination: when a model confidently says what isn't true · Level 3
2. Factuality and truthfulness: what counts as true, and how to check it · Level 3
3. Sycophancy · Level 3
4. Calibration: does stated confidence match correctness? · Level 4

## Sources

These lessons draw on 12 sources. The most used:

- [Survey of Hallucination in Natural Language Generation](https://arxiv.org/abs/2202.03629)
- [Language Models (Mostly) Know What They Know](https://arxiv.org/abs/2207.05221)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://arxiv.org/abs/2109.07958)
- [Why language models hallucinate](https://openai.com/index/why-language-models-hallucinate/)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://arxiv.org/abs/2212.09251)
- [International AI Safety Report 2026](https://internationalaisafetyreport.org/sites/default/files/2026-02/international-ai-safety-report-2026.pdf)
- [OpenAI GPT-5 System Card](https://arxiv.org/abs/2601.03267)
- [System Card: Claude Sonnet 4.5](https://www-cdn.anthropic.com/963373e433e489a87a10c823c52a0a013e9172dd.pdf)

[All units in Become an AI engineer](https://rabbithole.pro/courses/ai-engineer)

Source page: https://rabbithole.pro/courses/ai-engineer/truthfulness