In 2023, researchers started wondering if they could get away with only relying on AI-created data for training, instead of human-generated data.
Without high-quality human data, AI systems trained on AI-made data get dumber and dumber as each model learns from the previous one.
This 'regurgitive training' seems to lead to a reduction in the quality and diversity of model behaviour.
Quality here roughly means some combination of being helpful, harmless and honest.
Collection
[
|
...
]