python 3.10 doesn't let me use df.dt python 3.9 is ok

Showing a minimalist example below, I essentially have two functions, one to create a year column using "Date" column as input, and the other function which loads the data into a df from a csv file.

def create_year_column(df: pd.DataFrame) -> pd.DataFrame:    df['year'] = df['date'].dt.year.astype(str)    return dfdef load_data(path: str) -> pd.DataFrame:    # load the data from csv file    data = pd.read_csv(        path,        dtype={'amount': float,'category': str,'date': str,        },        parse_dates=['date']    )    create_year_column(data)    return data

The problem I have is that when I run this in python 3.10 I get the following error:

AttributeError: Can only use .dt accessor with datetimelike values. Did you mean: 'at'?

It seems to have issues with the line of code where I use:

df['year'] = df['date'].dt.year.astype(str)

I technically assume that the df['date'] column is allready parsed as a date format, which it is since its called in the load_data() function after I format it as such (using: parse_dates=['date'] ).

When I run this code in pyton 3.9 I have no issues, only 3.10 is giving me heartache. Anyone know why?