base 64 decoded not the same as original encoded value

Question

Ive created a login system that uses AES to encrypt user input. Below is the code that takes the input, pads, encrypts and then converts to base64. this is then stored in a csv which is also in the code below.

    #design padding function for encryption
    def padded_text(data_in):
        while len(data_in)% 16 != 0:
            data_in = data_in + b"0"
        return data_in


    mode = AES.MODE_CBC

    f_name_in = F_name.get().encode('utf8')
    S_name_in = S_name.get().encode('utf8')
    email_in = Email_22.get().encode('utf8')

    p = padded_text(f_name_in)
    p2 = padded_text(S_name_in)
    p3 = padded_text(email_in)
    print(p)
    print(len(p))

    cipher = AES.new(key, mode, IV=IV)

    fname_enc = cipher.encrypt(p)
    Sname_enc = cipher.encrypt(p2)
    email_enc = cipher.encrypt(p3)
    print(fname_enc)

    fname_enc64 = base64.b64encode(fname_enc)
    Sname_enc64 = base64.b64encode(Sname_enc)
    email_enc64 = base64.b64encode(email_enc)
    print(fname_enc64)
    print(len(fname_enc64))


    def save_to_csv():
        import pandas as pd
        import numpy as np
        import csv

        data = [[ID,fname_enc64,Sname_enc64,email_enc64]]
        header = ['ID','Fname','Sname','Email']

        df = pd.DataFrame(data)
        df.to_csv('enc_logins.csv', mode='a', index=False, header=None, encoding='utf8')
        with pd.option_context('display.max_rows', None, 'display.max_columns', None):
             print(df)

        dropbox_storage.dropbox_upload()
        Send_to_DB()

I then retrieve the file, import into a dataframe and base 64 decode. But this doesnt provide me with the same output as the original data before I base64 encoded.

original encrypted data via AES:

b'\x04\xe0l\xfe\x8b\xa5R, -Sl\xf2&\xf4\x9a'

above padded and encrypted to base64:

b'BOBs/oulUiwgLVNs8ib0mg=='

data after retrieving and decoding(correct and matches above):

b'BOBs/oulUiwgLVNs8ib0mg=='

result of decoding this from base64:

b'l\x13\x81\xb3\xfa.\x95H\xb0\x80\xb5M\xb3\xc8\x9b\xd2h'

What could be causing this incorrect base64 decoding? im being really stupid and missing something obvious im sure.

code for decoding below

    bs64 = lambda x: base64.b64decode(x)
    with open('enc_logins.csv', 'r', encoding='utf8') as csvfile:
        df = pd.read_csv(csvfile,  names=['ID', 'Fname', 'Sname', 'Email'],delimiter=",",  na_filter=False, encoding='utf8')
        newdf = df[df['ID'] == L_ID_entry.get()]
        print(newdf)

        df1 = newdf['Fname'].apply(bs64)
        df2 = newdf['Sname'].apply(bs64)
        df3 = newdf['Email'].apply(bs64)
        print(df1)
        print("df1 above")
    decoder()