I could tweak and run them to my liking.
I think for knowledge based things, you are better off using Google. But if it can help you narrow down your search, you can use that for Google.
For example, I queried, what projects are good for practicing data-intensive design.
It gave me a list of projects with descriptions. Perfect for me to Google more information for.
It's very good for narrowing down search and addressing follow ups. It's like the perfect Google Search companion.
Offers a chat feature that pairs the gpt-3 response with search results
If I search for python for loop syntax or metallica bob seger cover or tuck rule game year, etc. I know the right answer if you show it to me and given enough time to think I probably could recall it but I don't have the information at the top of my mind.
If chatgpt returns something that's wrong I'll know it's wrong and then maybe go check google.
This is exactly the scenario I encountered. I have a teammate who is using ChatGPT to ask questions as opposed to searching documentation and it gave incorrect info all around on all the following:
1. Where the property would be found in the event schema 2. What structure the value would be 3. The fact that it's not included at all for AWS AppSync specifically in question, even if the previous two had been answered correctly regarding API Gateway event schema.
All of this was available in the first doc returned in the search results of both API Gateway and AppSync respectively.
The fact it convincingly tells you otherwise causes additional friction since they're pointing at their screen telling me I'm wrong cause look at the response from ChatGPT.
Basically, if Google owned ChatGPT the result would be very different.
For many cases you can get enough information out of chatGPT to retrieve a specific corroborating reference. For coding tasks, you can just try the code, ask it for tests, or give it code review like feedback.
I’d suggest asking it about deep technical topics that you are familiar with, then work out it’s limitations. It’s effectively equivalent to a universal teaching assistant, sometimes wrong - usually points in the right direction.
It's waaaaay faster to get an immediate answer -- and be able to ask pointed, follow-up questions, which you couldn't otherwise do with a traditional site. And it's a lot easier to just fact-check statements online than to wade headfirst into something you know nothing about in random sites of varying qualities.
This augments ChatGPT results with Bing results to provide references you can look up.
Providing a list of 30,000 pages that are only tangentially related to what you want is not a virtue. -----
> Give me a link to page that gives an introduction to quantum computing.
Here is a link to a page that provides an introduction to quantum computing:
https://www.ibm.com/quantum-computing/learn/what-is-quantum-...
This page gives a brief overview of what quantum computing is, how it works, and some of the potential applications. It also includes links to more in-depth resources for those who want to learn more.
I hope this helps! Let me know if you have any other questions.
> Give me another link.
Here is another link that provides an introduction to quantum computing:
https://www.dwavesys.com/tutorials/getting-started-with-quan...
This page gives a high-level overview of quantum computing, including the basics of quantum mechanics and how it is used in computing. It also discusses the potential applications of quantum computing and some of the challenges and limitations of the technology.
I hope this helps! Let me know if you have any other questions.
...
A really specific example that has an affirmative, not just in a basic release but a leap forward: The beginning of 2022 is when Roborock released their robot vacuum dock that empties, refills, and cleans the mop on their top robot vacuum/mop combo.
So replacing SEO blog spam with ChatGPT is not an improvement. Blog spam optimises for SEO rules, ChatGPT optimises to sound like legitimate (but wrong) blog spam.
With ChatGTP you have to take the information on pure trust, and the information it gives is not at a level you can trust yet.
>> The Fourier transform is a mathematical technique that decomposes a function or signal into its constituent frequencies. It allows us to represent a function in terms of the frequencies that make it up, rather than in terms of its time or spatial behavior.
The Fourier transform has many applications in a variety of fields, including signal and image processing, engineering, and physics. It is a fundamental tool in the analysis of periodic phenomena and is widely used in the study of waves and oscillations.
For more information about the Fourier transform, you may find the following resources helpful:
Wikipedia: https://en.wikipedia.org/wiki/Fourier_transform
Khan Academy: https://www.khanacademy.org/math/differential-equations/lapl...
Wolfram MathWorld: http://mathworld.wolfram.com/FourierTransform.html
Problem solved.
OpenAI, specifically, has also been working on allowing GPT models to browse the internet and include citations with e.g. https://openai.com/blog/webgpt/
I don't know the actual statistics, but it seems like in my experience, at least 50% of the time, the linked sources are dead links, or don't have the information they're purported to have.
1. it actually has policy against this to make understanding sources simpler to the layman for any topic.
Its amazing at some things, but there are foundational mistakes everywhere. Wikipedia has always had much higher quality overall.
I seem it lack the capability to say it’s level of confidence in the anwers. I gave me very good result when I give him lots of context. You can put pages of code about a project and then ask a question to complete. In this case it’s way more accurate (more context)
At the speed it’s progressing I am sure it will be very good in a year or two.
How did the training of ChatGPT make it do this?
And could a different training avoid this?
ChatGPT, in my opinion, is great for "how do I code X" type questions, but isn't so good at the types of queries you mentioned, due to the lack of a search engine.
My weekend project was an open source combination of Google + GPT that returns pretty good results for these types of queries. You can check it out here - https://github.com/VikParuchuri/researcher
Example - the response to "what are the best current smartphones" is:
`...According to Search Result [2], the best phones have been thoroughly reviewed and tested, and include the Apple iPhone 14 and 14 Pro, the Pixel 7 Pro and the Samsung Galaxy S22 Ultra. Search Result [5] also states that there are strong options available at all price levels, so you don't have to spend a lot to get something great...`
[{“Key”:”Company”},{“Value”:”${company}”},…]
I asked it to write a Python script that replaces any word surrounded by ${} with the value in its corresponding environment variable and accept the path of the file as a command line argument —json-file using argparse. It worked perfectly.
Then I started asking it to write a script to successively do the following
Given the same json file, write a snippet of YML that looks like sample CF templates parameter section that I gave it.
https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGui...
It worked perfectly.
Then I told it to accept an optional argument that generated the corresponding meta data section. It worked flawlessly. I gave it sample expected output
https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGui...
Then I told it to output the format needed to pass those parameters to a nested stack and gave it an example;
Parameters: Company: !Ref company
…
It worked again.
Finally, I needed it to generate the Python code to generate the CodeBuild Environment section and I gave it an example.
https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGui...
`> askleo how large is the average dog`
`> runleo list files in reverse date order`