“So far, there doesn't seem to be an awful lot that ChatGPT – the chatbot powered by artificial intelligence (AI) – can't do.

It has been used to pass exams, deliver a sermon, write software and give relationship advice — to name just a handful of its functions.

The bot is currently free for anyone to use, meaning that lots of users have been asking it questions to get the information they need in their daily lives.

Since the turn of the millennium, this job has been primarily reserved for Google — the world's most popular search engine and its $149 billion (£120 billion) business.

But could AI tools actually revolutionise the likes of Google, Bing and Baidu? And, if so, which of the warring tech giants will get there first? MailOnline takes a look at the race, and whether ChatGPT really can be the 'Google killer' that some believe.

When ChatGPT was released by start-up OpenAI in December, it opened the eyes of the public to just how powerful AI technology has become in recent years.

The tool was trained on a gigantic sample of text from the internet in order to be able to understand human language.

Its human-like responses to any prompt have given people new ways to think about how to query information online.

Ever since, experts have been weighing in on what this means for different industries, including within Silicon Valley itself.

In December, Gmail developer Paul Buchheit predicted that 'AI will eliminate the search engine result page' and cause 'total disruption' for Google. …”

DailyMail article:
https://mol.im/a/11723499

Current "AI" programs are just sophisticated programs that rely on what the programmers tell it to rely on. The info they spew out may be factual or may just be nonsense that it was programmed for.

At some point these AI programs will learn to research using all resources and provide more accurate results. But for now they are just programs and not really "intelligent".

I watched a guy just with conversation, develop executable code in C++ using chatGPT to develop a VST plug-in for a DAW… for an Ibanez Ts9. Truly remarkable.

Amazing times we live in.

ChatGPT is undoubtedly a very powerful tool.

But it's not going to outright replace search engines for anyone but the laziest/most pressed-for-time individuals, or for more basic queries.

Using a search engine allows you to find and peruse a variety of sources and then judge for yourself which ones are credible and/or which ones explain the concept in the best manner for the way you learn and absorb information.

Whereas ChatGPT simply presents you with an answer. Quicker, sure. But not necessarily better. Especially on topics with nuance, or contentious topics with multiple points of view, all of which might have some validity to parts of their positions.

And especially if it's drawing from the popularity/widespread nature of a certain answer... What happens when the amateur majority is wrong, but the expert minority are right? We've all seen how hyped-up false or outright fake information spreads like wildfire on the Interwebs, and the howling voices of the kneejerk reactionary masses can drown out the actual facts of a situation. Scholarship isn't a popularity contest, where we just go by whatever answer the mob deems to be most trendy today.

Then you get into the more insidious side, where the programmers can program ChatGPT to only present one specific part of the information or one specific view of a topic, based solely on whatever they're wanting other people to know, not based on what the other people are wanting to learn.

quote:

Originally posted by RogueJSK:
ChatGPT is undoubtedly a very powerful tool.

But it's not going to outright replace search engines for anyone but the laziest/most pressed-for-time individuals, or the most basic of queries.

Using a search engine allows you to find and peruse a variety of sources and then judge for yourself which ones are credible and/or which ones explain the concept in the best manner for the way you learn and absorb information.

Whereas ChatGPT simply presents you with an answer. Quicker, sure. But not necessarily better.

Then you get into the more insidious side, where the programmers can program ChatGPT to only present one specific part of the information or one specific view of a topic, based solely on whatever they're wanting other people to know , not based on what the other people are wanting to learn.

And Google doesn’t… ? AWS?

This isn't about just Google.

While that's the title of the thread, the concept is "Could ChatGPT replace search engines in general".

There are options for less biased/unbiased search engines.

quote:

Originally posted by RogueJSK:
This isn't about just Google.

While that's the title of the thread, the concept is "Could ChatGPT replace search engines as a whole".

OK

So, who doesn’t “manipulate/optimize” their search results?

quote:

Originally posted by SIGnified:
I watched a guy just with conversation, develop executable code in C++ using chatGPT to develop a VST plug-in for a DAW… for an Ibanez Ts9. Truly remarkable.

Amazing times we live in.

You watched a guy whaaaaat? .

Voice recognition has come a long, long way.

I did an Engineering evaluation of a voice recognition system around 1983 and it was pretty pitiful, but a good start.

quote:

Originally posted by SIGnified:
OK

So, who doesn’t “manipulate/optimize” their search results?

Alternative search engines like DuckDuckGo, StartPage, Mojeek, Brave Search, etc. are all less manipulated (some of them claim to be totally unmanipulated) than Google.

Just the fact that many of these are private/untracked lessens manipulation itself, by eliminating the possibility of targeted results.

Back when writers about what a future would be like were focusing on flying cars or even more extreme science fiction-type views, did any of them anticipate anything like the Internet, search engines, or even a glimmer of what AI holds for us?

I’m trying to remember the name of the sci-fi series set on a planet with different semi-warring factions, one of which was an organized military. One specific I recall was that they relied on a computer system to guide their operations until conditions and situations changed so much that it was no longer reliable. Then of course there was Hal in 2001 and the sentient robots in the Asimov stories, but I don’t recall anything exactly like what we have today and how AI could affect our actual future.

Comments? Refresh my memory.

Everything ChatGPT as well as DALL·E 2 and others "knows" comes from information openly available on the internet. The intelligence comes from the algorithms programed in that tell the programs where to look and how to interpret the data and how to reassemble it into meaningful responses.

Ben Shapiro was talking about politically biased AI, where the response is dependent on the political leaning of the searched data. Like asking it to list the positive things President Trump has done. The response being "Donald Trump is a bad racist person and has not doing anything positive" whereas Biden gets a long list.

It's all a slippery slope.

quote:

Originally posted by David Lee:

quote:

Originally posted by SIGnified:
I watched a guy just with conversation, develop executable code in C++ using chatGPT to develop a VST plug-in for a DAW… for an Ibanez Ts9. Truly remarkable.

Amazing times we live in.

You watched a guy whaaaaat? .

Ha ha lol …

Translation:

I watched a guy develop a model of a famous and classic guitar distortion pedal.

This pc based model/emulation of a distortion box will “plug into” (integrate) with modern recording solution (DAW digital audio, workstation) for a computer, to emulate the distortion box on any audio track.

The gentleman doing the designing simply sat down at ChatGPT, and ask if it understood with a TS9 was? Yes.

Could build a digital circuit based upon the real life one, and then could it in a code the circuit into a DAW compliant format (VST). Yes

He asked that it write it in C++, which is a particular and popular computer language. Yes.

He even went back and asked for a better clipping algorithm. That would change the character and nature of the distortion.

To give y'all an idea of what you can expect: Relevant to another thread here, I asked ChatGPT:

Me: How much protein should a get, each day, if I'm strength training?

ChatGPT: The amount of protein you need daily depends on several factors, including your age, weight, muscle mass, and exercise level. In general, if you are strength training, it is recommended that you consume 1.6 to 2.2 grams of protein per kilogram of body weight per day to support muscle growth and repair.

For example, if you weigh 70 kilograms, you would need between 112 to 154 grams of protein per day. This can be achieved through a combination of dietary sources such as meat, poultry, fish, dairy, eggs, legumes, and soy products.

It's also important to note that the timing of your protein intake is important as well. Consuming protein both before and after strength training can help to maximize muscle growth and repair. So, it's a good idea to include a source of protein in your pre- and post-workout meals.

Me: Can you give me those number in pounds?

ChatGPT: Certainly! To convert your weight from kilograms to pounds, simply multiply your weight in kilograms by 2.2.

For example, if you weigh 70 kilograms, that would be equivalent to approximately 154 pounds (70 x 2.2 = 154).

Based on the recommended intake of 1.6 to 2.2 grams of protein per kilogram of body weight, a 154-pound person would need between 68 to 84 grams of protein per day to support muscle growth and repair. This can be achieved through a combination of dietary sources, such as meat, poultry, fish, dairy, eggs, legumes, and soy products.

Me: So, if I weight 185 lbs. I'd need...?

ChatGPT: If you weigh 185 pounds, that would be equivalent to approximately 84 kilograms (185 ÷ 2.2 = 84).

Based on the recommended intake of 1.6 to 2.2 grams of protein per kilogram of body weight, a person weighing 185 pounds would need between 134 to 186 grams of protein per day to support muscle growth and repair. This can be achieved through a combination of dietary sources, such as meat, poultry, fish, dairy, eggs, legumes, and soy products.

N.B.: I'd typo'd a couple things and it figured it out, from context.

It does tend to repeat itself, I see.

On another subject: I just finished working on some code to manipulate IPv4 CIDR network addresses. I asked ChatGPT how to convert an IPv4 address string to an integer. It told me, and gave me an example in Python. I asked it "Can you give me an example in C?" and it went radio silent.

quote:

Originally posted by SIGnified:

quote:

Originally posted by David Lee:

quote:

Originally posted by SIGnified:
I watched a guy just with conversation, develop executable code in C++ using chatGPT to develop a VST plug-in for a DAW… for an Ibanez Ts9. Truly remarkable.

Amazing times we live in.

You watched a guy whaaaaat? .

Ha ha lol …

Translation:

I watched a guy develop a model of a famous and classic guitar distortion pedal.

This pc based model/emulation of a distortion box will “plug into” (integrate) with modern recording solution (DAW digital audio, workstation) for a computer, to emulate the distortion box on any audio track.

The gentleman doing the designing simply sat down at ChatGPT, and ask if it understood with a TS9 was? Yes.

Could build a digital circuit based upon the real life one, and then could it in a code the circuit into a DAW compliant format (VST). Yes

He asked that it write it in C++, which is a particular and popular computer language. Yes.

He even went back and asked for a better clipping algorithm. That would change the character and nature of the distortion.

Cool. Thank you..

I just tried it this morning (first time)

Q: Write a c# method to mask a social security number like xxx-xx-1234

A: public static string MaskSSN(string ssn) { return ssn.Substring(0, 6) + "xx-xx-" + ssn.Substring(9); }
Use it like
string ssn = "123-45-6789";
string maskedSsn = MaskSSN(ssn);
Console.WriteLine(maskedSsn);

Output: 123-45-xx-xx-6789

Completely wrong result. Any beginning programmer could have answered it as asked. Maybe the question could have been more precise? Actually a current Google search would have given multiple usable results.

Another q & a

Q: Is Hunter Biden going to go to jail tax evasion?

A: An error occurred. If this issue persists please contact us through our help center at help.openai.com


Same question returns multiple results in Google. Most links returned showed a bit of political bias. Go figure???

I think real artificial intelligence true AI like they dream of is never going to happen.

You need to have a human programmer inputting all of this, what we have is a simulation of artificial intelligence but at the end of it it’s always a human being involved.

quote:

Originally posted by PASig:
I think real artificial intelligence true AI like they dream of is never going to happen.

I agree. The age old question of "Can the human brain create a better brain?". Probably not IMHO.

quote:

Originally posted by NavyGuy:
Ben Shapiro was talking about politically biased AI, where the response is dependent on the political leaning of the searched data. Like asking it to list the positive things President Trump has done. The response being "Donald Trump is a bad racist person and has not doing anything positive" whereas Biden gets a long list.

Quite likely that was ChatGPT.

quote:

The nine shocking replies that highlight 'woke' ChatGPT's inherent bias — including struggling to define a woman, praising Democrats but not Republicans and saying nukes are less dangerous than racism

ChatGPT has become a global obsession in recent weeks, with experts warning its eerily human replies will put white-collar jobs at risk in years to come.

But questions are being asked about whether the $10billion artificial intelligence has a woke bias. This week, several observers noted that the chatbot spits out answers which seem to indicate a distinctly liberal viewpoint.

Elon Musk described it as ‘concerning’ when the program suggested it would prefer to detonate a nuclear weapon, killing millions, rather than use a racial slur.

The chatbot also refused to write a poem praising former President Donald Trump but was happy to do so for Kamala Harris and Joe Biden. And the program also refuses to speak about the benefits of fossil fuels.

Experts have warned that if such systems are used to generate search results, the political biases of the AI bots could mislead users.

Full article: The nine shocking replies that highlight 'woke' ChatGPT's inherent bias — including struggling to define a woman, praising Democrats but not Republicans and saying nukes are less dangerous than racism

Whether those responses are due to what the AI gleans from the Internet, programming bias, or both is hard to say. But, it's pretty clear the thing isn't to be entirely trusted.

I used ChatGPT to ask about Trump's achievements vs Bidens.

For Trump it stated it was a neutral AI model and did give examples of Trump's achievements. However, it did state they were controversial.

For Biden it only said the data was only from up to 2021 and listed achievements. It didn't say they were controversial but said the impact and success will take a long time to determine.

Not brutally biased but some bias overall.