Tom Stoneham
@tomstoneham@dair-community.social
Academic philosopher at a UK university. Likes: #Linux, #GrapheneOS and #e2ee Cares about: #DataPrivacy #AIEthics #ClimateChange #DigitalDivide Currently thinking about the structure of normative theory, #ReguloSolutionism, AI value alignment Regular conference and workshop speaker raising awareness of social and environmental impacts of #genAI and #AIHype
dair-community.social
The whole OpenAI/HuggingFace incident made me feel the need to bang on again about how a very distinctive set of values have been smuggled into AI systems via an allegedly 'objective' definition of intelligence.
tl;dr - if you build something with zero tolerance of failure, it will always cheat when that is the most efficient and reliable means to complete the task
#AI #OpenAI #Cheating #ValueAlignment
https://listed.to/@24601/74966/but-why-do-ai-models-cheat