> wouldn't this incentivize confidently guessing (making stuff up) Making stuff up is not the same as honestly making statements that you believe are correct,

> wouldn't this incentivize confidently guessing (making stuff up) Making stuff up is not the same as honestly making statements that you believe are correct, but which are actually wrong. In the latter case, the person (Pichai in this case) could give reasons why he held that belief (for example, so and so gave him a briefing on the feature and he understood them to be telling him X). In the former case (guessing), he wouldn't; he would have to admit under oath that he just made the stuff up. I don't know if that technically counts as perjury, but it certainly doesn't seem to be consistent with giving testimony under oath.