Gary Marcus
Deep Research is genuinely useful - depending on your application - but crucially (as anticipated by Rebooting AI in 2019, and by @yudapearl) facts and temporal reasoning remain problematic for current neural network-based approaches that lean heavily on statistics rather than deep understanding.
Sayash Kapoor
@random_walker @colin_fraser @emollick Overall, Deep Research is impressive. But it is not a replacement for all research assistance. It is better for some uses (search, case studies) than others (factual, temporal questions). How easily we can fix the limitations of factuality is an important open question.