Optimistic service validation vs Pessimistic validation

Question

imagine that you have to create a method in a ClientService class that you pass in the clientId and this method returns a ClientDTO. I have seen many times that this service methods does not validate for example if the clientId exists. I show you two versions:

**//Version 1(optimistic without validation)**
public ClientDTO getClientData(Long clientId){
   ClientEntity clientEntity = clientRepository.findById(clientId);
   return new ClientDTO(clientEntity);    
}




**//Version 2(pessimistic with validation)**
   public ClientDTO getClientData(Long clientId){
       ClientEntity clientEntity = clientRepository.findById(clientId);
       if(clientEntity  == null)      {
          throw new IllegalArgumentException("Client id not found");
       }
       return new ClientDTO(clientEntity);    
    }

Questions:

1. Is there some kind of optimistic/pessimistic paradigm in programming when we talk about validation?
2. I have been said that for example the version 1 is OK because previously to call this method, some other service has validated that this clientId exists. I think this way of thinking is wrong because this method has to be agnostic from who is calling it and therefore does not have to know that the clientId has been validated before in other service. What do you think?.
3. Are there some cases where version 1 applies better than version 2?