Hacker News new | past | comments | ask | show | jobs | submit login

> There are blackout periods etc. just to make sure that the analysis is valid and not tainted.

This is an interesting potential debate. In some sense we are sacrificing lives for scientific validity, because it is potentially possible that we could deploy early, ignore blackouts, and have probabilistically fewer deaths + tainted data. But the world is apparently wanting to choose probabilistically more deaths + clean data.




> In some sense we are sacrificing lives for scientific validity, because it is potentially possible that we could deploy early, ignore blackouts, and have probabilistically fewer deaths + tainted data

This is spurious logic.

How are you saying that tainted science + bad data will result in fewer deaths long term?

You may end up with probabilistically more deaths + tainted data.

Why is that not a possibility?




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: