Though manual testing gives us more accuracy to find out the critical issues but for some cases we have to depend on automated testing to find out few simple issues, which cannot be found by manual testing. Here is a simple example:
In one of our project, we have a field called Tags in content create page. For this field it was the requirement that users should not get the permission to add more than five (5) tags. It was implemented using JS and I have manually tested it and found ok. But while I have re-checked it using Selenium IDE then I was able to add more than five (5) tags. Most important thing is that I have not only added the tags but also submited the form. So if users want, they can add huge numbers of tags using automated test scripts.
Actually here the test scripts has bypassed the JS and broken the functionality. So we should be careful about this type of implementation and not to forget about automated testing.