en.wikipedia.org/wiki/Regression_toward_the_mean

This overstates Galton’s role. Francis Galton’s work on heredity was historically important in naming and popularizing regression toward the mean, but the underlying linear least-squares regression methods predate him by many decades.

A National Institute of Standards and Technology (NIST) overview states that the method of least squares used for linear regression parameter estimation was developed in the late 1700s and early 1800s by Gauss, Legendre, and possibly Robert Adrain. A Utah State University history page likewise explains that scientists were fitting bivariate data with least squares regression in the early 19th century; it specifically names Legendre (1805), Adrain (1808/1809), and Gauss (1809), and notes that Laplace also used linear regression methods before Galton. The same page then describes Galton’s contribution as a further development and application of regression, not its invention.

So the accurate historical claim is that Galton coined and popularized the term “regression” in this context, but he did not invent linear regression analysis itself.