It links to https://www.iihs.org/topics/fatality-statistics/detail/yearl... for that 7500 pedestrians in 2022 figure.
Yet on that very same page there is a table of pedestrian and bicyclist deaths going back to 1975, which shows that 1977-1981 all had more pedestrian deaths (7732, 7795, 8096, 8070, and 7837) than 2022 (7522). Several of those years are within the last 45 years.
For bicyclists 2022 is indeed the biggest year just edging out 1977 (1084 vs 1003).
This might be forgiven as just being a little sloppy. If he'd just have said "past 40 years" instead of "past 45 years" then it would have been accurate.
But wait...the US population was 338 million in 2022, and around 220 million in 1979. We really should look at rates.
2020 pedestrian deaths per 100k population were 2.22. They were above that every year from 1975-1991. Same for bicyclists--every year from 1975-1991 was higher than 2020.
For pedestrians the rates peaked in 1979 at 3.76 then declined fairly linearly to 1.33 in 2009, then rose again fairly linearly to 1.95 in 2020, then jumped a bit to 2.22 for 2021 and 2022. Same pattern for bicyclists.
Here's a graph of the rates by year [1]. The bicyclist rates are per 1m instead of per 100k to make them easier to read.