'Current LLMs introduce substantial errors when editing work documents': Microsoft scientists find most AI models struggle with long-running tasks — so maybe don't trust them completely just yet | Collector
TechRadar
'Current LLMs introduce substantial errors when editing work documents': Microsoft scientists find most AI models struggle with long-running tasks — so maybe don't trust them completely just yet
The more interactions an AI model has, the less reliable it becomes, experts find, as even the best only scored 80.9% – and the worst scoring just 10.0%.
TechRadar