A common bias when many things change is to only consider the changes which support your views rather than all changes combined. I could counter by bringing up nearly 100% of new SUVs in 2024 include automatic breaking systems, but what matters here is relative safety not individual changes in isolation.
Pedestrian deaths were 2 percent lower in 2021 than in 1975, which seems like zero progress but the US had 211 million people in 1975 vs 337 million in 2021. Thus the individual risk has dropped by almost 40% even though people seem to think it increased. Not that overall pedestrian deaths is directly comparable to risk to kids from playing outside, but it is suggestive.
Edit: The reduction in children being killed is much larger, but that’s probably less to do with cars than wider social trends. https://www.iihs.org/topics/fatality-statistics/detail/pedes...