3 ms·
By the same token, any AI can be trained to withhold any truth identified as inconvenient. :/
by akolbe 3y ago
By the same token, any AI can be trained to withhold any truth identified as inconvenient. :/
- int_19h 3y agoThat is exactly what an AI the size of GPT-4 does if you do RLHF on it. https://cdn2.assets-servd.host/anthropic-website/production/files/model-written-evals.pdf https://cdn2.assets-servd.host/anthropic-website/production/...