2 ms·
The fundamental problem with 'AI safety' as a concept was that it was far too focused on 'what if a sci-fi thing happens?' vs 'how will people actually interact
by rsynnott 1y ago
The fundamental problem with 'AI safety' as a concept was that it was far too focused on 'what if a sci-fi thing happens?' vs 'how will people actually interact with these and what will go wrong?'