ChatGPT is not basically Google, because ChatGPT is not a search engine.
Google works by querying billions of sources to find the most applicable/useful/reputable ones.
ChatGPT works like the “hint words” when messaging someone. It just guesses what a sentence should look like, given the prompts it has seen.
It is often surprisingly accurate, mostly because it has been trained on a similar set of billions of sources. But all it is ever doing is predicting which words, put together which way, would appear to be a valid response to your prompt.
It’s not google because it’s *generative*. It takes the words and metadata and all of the stuff on the internet, and it can make something new based off of that.
The best example is the [lawyers that used chat-gpt](https://www.forbes.com/sites/mollybohannon/2023/06/08/lawyer-used-chatgpt-in-court-and-cited-fake-cases-a-judge-is-considering-sanctions/?sh=24a45c847c7f) and it spit out fake cases that never actually happened. Google search cannot generate things like that, it can only show what was generated by other people/bots.
This property can be bad, like in the lawyer case, or it can be good, like if you need a specific code snippet on an obscure subject, it can give you custom code rather than you having to search, parse the stack overflow posts, see if it’s actually what you need, and then adjust it to what you need.
And people are just being scared of doomsday scenarios, just like they always have. It’s nothing new to be scared of something that is new.
With a few exceptions, like weather and simple math, Google searches simply return links to web pages. ChatGPT returns information directly. Ask Google to write a cover letter and you’ll get a link to a web page about cover letters. Ask ChatGPT and it will write you the cover letter.
Google will return you links to a poem about unicorns. ChatGPT will write you a brand new poem instead.
Latest Answers