I know what you mean! But here's the thing.
I will start “from afar” - have you ever noticed that when you ask a question, for example ChatGPT or Grok, in the answer window runs quickly something like “Page search”. And it's not just for fun
In fact
ChatGPT/Grokand some other systems can work in 2 modes:
Give answers based on a KNOWLEDGE base and a large language model
Give answers by processing through a search engine, a few selections of relevant sites, finding the necessary data, which allows you to work with “the most relevant data”. For tasks like “what is the score”, when you don't need to think and just take information from open sources - the second mechanism works. I am not ready to say whether these data are used as raw data for training. But it's not “AI”, it's just data mining, like Google and other search engines.
But I use it 99.9% for decisions where knowledge, logic, and the ability to “think” are needed. Agree - there is a difference between thinking about solving a complex problem (e.g. writing code) and searching for information in open sources !
Try to load them with some specific, narrow-profile tasks, the solution of which must be thought out ! And there the quality of answers will be different.
I will start “from afar” - have you ever noticed that when you ask a question, for example ChatGPT or Grok, in the answer window runs quickly something like “Page search”. And it's not just for fun
In fact
ChatGPT/Grokand some other systems can work in 2 modes:
Give answers based on a KNOWLEDGE base and a large language model
Give answers by processing through a search engine, a few selections of relevant sites, finding the necessary data, which allows you to work with “the most relevant data”. For tasks like “what is the score”, when you don't need to think and just take information from open sources - the second mechanism works. I am not ready to say whether these data are used as raw data for training. But it's not “AI”, it's just data mining, like Google and other search engines.
But I use it 99.9% for decisions where knowledge, logic, and the ability to “think” are needed. Agree - there is a difference between thinking about solving a complex problem (e.g. writing code) and searching for information in open sources !
Try to load them with some specific, narrow-profile tasks, the solution of which must be thought out ! And there the quality of answers will be different.
So far I have used ChatGPT more then any AI chat bot available in the market like DeepSeek or Grok. I agree that there are scenarios where ChatGPT gives you answer by simply coping text from websites on the internet and this is like saving your time to search online.
From coding perspective, it gives you code that works but it can't be said that the code generated by ChatGPT is 100% according to your requirement. If you know basic of coding then you can fine tune it with little effort. Overall it saves lots of time that you will spend in writing the code from scratch.
One draw back of these AI chat bot is that students are not doing any effort in doing there school(college) homework(assignments), they are just giving the questions to these Chatbots and pasting the answer with some amendments to avoid AI detection.
As a developer with a long experience, I will say this - for me such systems are useful in such directions:
1. generation of primitive but very voluminous code or similar information
2. Searching for specific errors in large projects (he is a bit more attentive than me
)3. quick answer to a question that I can search for myself, but... I save time. But then I make sure I understand the topic thoroughly - because in the end I should have knowledge.