SSSufe says they need the line with the largest number, not the longest line, so to expand on 1nickt's solution:

#/usr/bin/perl
use strict;
use warnings;
use feature 'say';

my $lineWithMax;
my $max = 0;
my @numbers;
my @lines = <DATA>;

for my $line (@lines) {
    chomp($line);
    @numbers = $line =~ /(\d+)/g;
    foreach my $number (@numbers) {
        if ( $number > $max ) {
            $lineWithMax = $line;
            $max = $number;
        }
    }
}

say "Line with highest number is '$lineWithMax'";

__DATA__
Hello, i'm 18

1 this year is 2019 1

1 2 3 - 4
[download]

The code works with your sample data, but consider the following: If two or more lines contain the same highest number, it will output only the first one. If a line contains a negative number, it will be treated as a positive one (the "-" will be ignored). Only integers are considered. Numbers written with scientific notation, for example, will not be treated correctly (4 x 10⁵ would be considered less than 40) and decimals would be truncated and the remaining digits treated as a new number.

Today I learned that captured groupings from a regex can be put in an array.