About whether any task really needs a humanoid robot rather than a specialized one.
Has this happened?
Yes 2 (2%)Not sure 10 (8%)No 119 (91%)
Votes cast 1–2 October 2026: 100,590 votes from 9,694 people.
Hacker News has set AI a lot of challenges over the years. Which ones has it met?
About whether any task really needs a humanoid robot rather than a specialized one.
Votes cast 1–2 October 2026: 100,590 votes from 9,694 people.
The case I have in mind: cleaning hotel rooms. Build me a robot that can clean the toilet, change the sheets, replace the soap, check the mini-fridge, and do it all without breaking people's stuff scattered around the room.
This is minimum-wage work which probably costs you $30,000 a year (including taxes, benefits, etc.) A $100,000 robot would replace that cost in just over three years -- one year, if it could work 24 hours a day instead of 8.
That seems a huge ask. The tasks are all simple, but currently require AI-complete levels of perception, especially if combined into one machine. I doubt that a specialized robot -- or a series of them -- is likely to be cost-effective. Currently, they don't even use roombas.