Google Panics Over ChatGPT [The AI Wars Have Begun]

Google's newly announced BARD AI system is mentioned at 12:25
In this episode we see why Google has called a code red because of ChatGPT but why? Why is ChatGPT such a threat? And what are Google doing about it? In this episode we find out.
Animations by:
/ @thenwhathappens
ColdFusion Podcast:
/ @throughtheweb
ColdFusion Music:
/ @burnwatermusic7421
burnwater.bandc...
Last Track:
• Burn Water - Don't Eve...
Previous AI video:
• It’s Time to Pay Atten...
Get my book:
bit.ly/NewThink...
ColdFusion Socials:
/ discord
/ coldfusiontv
/ coldfusion_tv
/ coldfusiontv
Producer: Dagogo Altraide
Writer: Meehan Kathan and Dagogo Altraide
Editor: Tanzim Uddin and Dagogo Altraide

[@hazonku]

THAT'S IT! This whole time during Chat GPT's rise I've been asking myself, "Where the Hell is Google? Haven't they been leading in AI academic research for a solid decade+ now?" And the answer was obvious and right in front of us the whole time. The answer was ads. Google isn't a search engine company, it's an ad broker. Their business isn't search, search is just how they get you to the ads & how they collect information on you to sell to advertisers. If Google made the search engine AI driven you won't open 5 different sites looking for that oat milk recipe and you'll never see the dozen or more ads & popups on each site.

[@annoloki]

But if you never open those 5 different sites, and neither does anybody else, they wouldn't exist. (I don't mean in a "tree falling when nobody's around" kind of way, I just mean that nobody paying for the sites will mean they don't remain online!)

[@bogmasher]

@@annoloki great, get rid of all the crap, quality over quantity every time!!

[@TommyCubed]

@@annoloki Ever heard of "Dead Internet Theory"?

[@DualDGamer]

@@annoloki if everyone uses chatgpt or another for the bulk of that stuff, then the bulk of that stuff won't remain online. Kind of ironic that what was needed to create chatgpt will be destroyed by it.

[@ak15567]

I guarantee people smarter than you and I will find ways to shove ads down our throats. I wouldn't be surprised if they charged a premium for these kinds of responses in a monthly subscription or an ad version. I think the main point for the slow adaption was that these AI can be biased and unfiltered. Google would never hear the end of it if a response was anti-conversative and they'll be hearing it out on committees for weeks or months for every controversial response that the AI generated.

[@richard2218188]

I think the biggest challenge to Google right now is that, people may believe "there's an alternative"

[@hateeternalmaver]

Yeah it's all in how the companies play their cards now. It's not like there wasn't ever any (even better) alternatives but waking up the masses from their paralysis sleep and grabbing their attention for a moment is really a huge deal it seems.

[@markpoidvin5382]

I think all search engines miss why Google is a success. The results are basically the same with all. The page is clean. That's it. Bing tries to sell you everything under the sun. They have 7% of searches, and 90% of that is because you have not turned off all the default searches in Windows that they don't make easy.

[@bladezx4880]

They've been pushing the agenda of one side instead of making the world a better place. That's on them.

[@UCjNrKLyRJI-abFA8qiNo92Q]

we've been using duck for years now

[@john_avernia]

A lot of people believe that. Just like the person above me, me and my siblings been using duck duck go for our search. And now, I use chatgpt and duck.

[@TimeBucks]

As always another comprehensive video.

[@subhashchandra2008]

Great

[@victormogoarias]

oool

[@dimasasongslowedreverb4724]

Nice

[@AsifShah-mt8ei]

Fantastic

[@AsifShah-mt8ei]

Fantastic

[@matthewharris517]

One thing I always HATED about Google was their oversaturation of ads

[@leroybrown505]

or the prepicked first 5 pages. The condition people with their picked pages. I miss the old internet . Google needs to go bankrupt.

[@LGoromonzi]

This is not a good sign, when you refer to Google in the past tense........🥶🥶🥶🥶🥶🥶🥶🥶🥶🥶🥶🥶 Is winter here already?😰

[@nonshatter7]

As much as I also dislike ads, we are getting their stiff for free and it has to be paid for somehow. I just wish they would improve their targeted ads; items I would not mind watching. I have seen some good ads but they are few a far between.

[@Epoxydeable]

That what ones on the top do

[@justinruiz5054]

On one hand, there will me more worse ads than ever. On the other hand, ai can be made to be an ad blocker.

[@Nil-js4bf]

Things are definitely getting shaken up. I didn't realize the impact this would have on ad revenue. ChatGPT synthesizing the full answer for you would mean less clicks to the source sites. And that's a massive issue when your most of your income comes from ad revenue which itself depends on getting user clicks. No wonder Google hasn't released something like this earlier even though they should already have all the data, talent and resources to do so. It would be disrupting their own business model when they already dominate the search industry.

[@MichaelSinz]

This is exactly the innovator's dilemma: Don't disrupt your own "cash cow" and you allow someone else to do so. Not saying that this was the main reason but ... Then again Google still has the "I'm feeling lucky" button

[@puddintime8107]

Well said. Just like Gm had electric cars in the 90s and then pretended no one wanted them

[@commentfreely5443]

microsoft will be the same as google with extreme far left talking points controlling every search

[@MichaelSinz]

@@commentfreely5443 just facts controlling the replies. If you think the earth is flat or vaccines don't work, you will be disappointed with the search results as they will return what the facts are. If you think the facts are left talking points then you need to re-examine your basis for thinking that.

[@alden9086]

@@commentfreely5443 Agreed. God forbid you simply ask the bot to generate a Devil Advocate's argument. Something being "horrible" or "morally irrehensible" are opinions that limits someon's ability to be objective. Having your views and opinions challenged in a thought-provoking manner is very important, in my opinion.

[@randomcharacter6501]

Google had the perfect algorithm back in the late 2000's. Then they decided to know what you want better than you do, also the lack of resource web pages... Replaced by SEO and other companies. No matter what you search now you'll get something sold to you for the first few results. They deserve to be worried, even tho Chat GPT is using a different tech, they're just redoing what Google was already doing over 10 years ago.

[@atticusherodes6648]

And the SEO matters less and less. And is overridden manually. And replaced with either something rewrite or replaced with the fake enclyopedia.

[@rapid___]

Why do people use SEO like some sort of evil tech? It's just people making their website easier to find. It's not something done just by big companies - I've done it on my personal websites and so have friends and family so that when someone looks us up they find our personal sites before our Facebook or whatever. SEO makes things more discoverable, and the search results more relevant, not worse.

[@torturedexistence029]

@@rapid___ it's a double edged sword. do a simple google search on a topic and you'll be flooded by the same articles written and rephrased over and over again on different websites just to create traffic. most of the times, these articles are outdated or mostly aren't even what you need. nowadays its easier and quicker to find what you're looking for in reddit comments

[@madjack730]

I don’t like google anymore. Reddit is better. Google is full of to many ada

[@nuthinnew]

Reminds me of McDonalds. That film, The Founder, shows how they took their menu down to just a few items so there was no wait time in the queues which is probably how they became popular. Nowadays you are waiting 10 minutes for it

[@YourBlackLocal]

Honestly Google’s search engine has been getting worse and worse.

[@integralogic]

It's basically a propa Ganda search engine. We are turning into china

[@FocconMc]

Yep, the censorship is real.

[@vissermatt1058]

not sure if "worse" is the correct word. it really isnt open and balanced, and hasnt been for a long while. they pick and choose what region websites you cans see, and they defintely make money off of prio rules. business descisons over search engine

[@linklanre]

It's because Google shows results that had been optimized to land on Google first page. Nonsense blog post could be optimized to appear on Google first page of search.

[@patrickpaterson8785]

@@vissermatt1058 That just sounds like worse with extra steps

[@jonathansl109]

It's crazy how long the voice assistant have existed and how they have just stagnated, nearly 10 years ago I remember using the Google voice assistant and it was crappy but since it was new... And now it's as crappy as ever, Siri is even older... Really mind-blowing to think how little those products have improved.

[@Akotski-ys9rr]

I remember thinking the google voice assistant was mind blowing when I saw it in advertisements

[@amelliamendel2227]

They haven't had a way to monetize it so they didn't put any money into it themselves apparently.

[@NatCatKitty]

Imagine if Siri was as smart as chat GPT that would be pretty cool

[@Naokarma]

@@NatCatKitty Honestly, imagining Siri/Google VA being as smart as my washing machine feels difficult at times.

[@NatCatKitty]

@@Naokarma haha that’s true

[@Madeleine_des_Esseintes]

I'm pretty disturbed by the idea of a search engine that doesn't only determine which sites you see first (as Google does) but that doesn't let you see any sites and sources and that doesn't let you roam (relatively) freely yourself. There's so much room for manipulation. It's like entering and traversing a (carefully curated) library vs. being forced to stand at the door and wait for the librarian to hand you a book of their own choosing.

[@youguessedit995]

And sometimes the book that the librarian hands you is pure bullshit.

[@Rockmaster867]

Well, they could add a "show sources" button that links all source web sites. Or you could just add it in your query

[@forthekozofhelpingothers]

In my opinion, I'm not a fan of the way Google is operating today. It was better a few years ago. Now, not only have I continued to have issues and have been hacked, so have several others. I don't like the fact that I am forced to have my privacy, data, etc violated. No matter what I choose for privacy settings, it's still being done. I have a history of stalker, DV, being assaulted. I do not want to take any chances for my information to be public and be in danger as well as my kids. But, I have tried to stop etc. Now I have to spend about 10 to 15hours week checking to see what is now public and go through the ridiculous hoops to attempt to get removed. This is not how I should be spending my time as it's the 3rd time of severe cancer and medical issues. Not sure if I will be able to make it through this time. Should be creating memories instead of all the crap I have to do now. It would be nice if they actually cared still about the user instead of the money being made from user data.

[@Flyingjaffacake]

And the book they hand you is written on the fly by committee based on what it thinks you want to know.

[@nathan8659]

I think that was certainly the case for a long time, but I don't believe the "library of Google" is as carefully curated as I would like.. I still use it by default, but there's definitely a chance that the "librarian" of Microsoft's new search engine will hand me the book I want sooner than I could find it wading through Googles shelves and checking the credibility of each book.

[@NickCombs]

The issue with GPTs in this context especially is the lack of transparency in the sourcing of data. Without that, it will be hard to trust their conversational answers to searches. Compare that to wikipedia and the limitation becomes apparent. So what we need is a GPT that shows its homework so to speak.

[@crogers2009]

It will also just make sources up. I asked for a paragraph on something, asked for the sources, and gave me links to websites and titles of articles that just don't even exist.

[@samuelmarndi]

with time it can become so accurate that we won't need proof of sources. except for news like content.

[@rochsmatahari7828]

You just need to ask chatGPT to provide the source ; book, ebook or the website.

[@Lanarri]

It would probably be nice to see a versions of algorythm trained only on limited and „proven” sources of data, such as Wki-only trained version of chatGPT. Probably more trusted sources would be added later on.

[@Lanarri]

And, oh damn, it would have to be human-generated trusted siurces, not AI based.

[@ross4]

“We’re building AI for the future benefit of humanity, not for corporate profit.” *Immediately sells out to Big Tech at the first opportunity.*

[@danielgroenewald69]

They had me in the first half though ngl

[@capslock9031]

This. Fucking dystopic, once again.

[@maebunny5298]

Who are you referring to? Who sold out to who?

[@pilhei]

Yeah, It's very naive to think that chatGPT will not shove ads after becoming as big as google.

[@pasta-and-heroin]

@@maebunny5298 the creator of chatGPT initially started the company as a non-profit, however this changed (i want to say in like 2014? google it lol) when they sold to microsoft

[@velosofy]

A question that comes to mind is that if all results of search engines will be based on historical data and ai generated data. Would this mean the death of content creators that write the articles and blogs where this data is originally from? How would this be sustainable in the future if no new content will be written anymore since AI has taken over here.

[@KnightMirkoYo]

There will always be space for cutting-edge research and writing. Just the more mundane summarization of what was previously said hundreds of times will mostly get automated. But it doesn't mean people will stop having blogs! Plus, since these are pre-trained models, they won't be able to quickly write about the most immediate events and news.

[@McDonaldsCalifornia]

That really shines a light in who is doing the actual work that go into these AIs. Ultimately they are just machines that extract surplus value from immaterial labor better than any machine before. Like an assembly line or a factory Floor but for language and knowledge creation. And in classic capitalist Fashion the owners of the capital get to sell us back the fruits of our collective labor at a profit

[@melodywong8265]

Yes, does it mean eventually most content will be written by AI, then summarised and regurgitated by AI, with no human content creation involved. Will knowledge become centralised into the few major AI bots, and then knowledge will be controlled by those companies and interests that control them?

[@electronresonator8882]

AI doesn't give you a perfect answer, and sometimes it gives you a wrong answer, so people will seek credible source to satisfy their curiosity, but since most people rarely clarify what they read in social media feed and just re-broadcast it again, so content creators will become much much smaller, only those who have valid and credible answer can survive, since only very few people bother to search that

[@shreyaszare1559]

Content on the web is already partly dead due to SEO garbage.

[@Vranabg]

this will kill most websites and cause even greater information control by corporate entities.

[@insightcentral]

Thank you. I don’t know why nobody is talking about this. Having multiple websites promoted decentralization of info.

[@swaggitypigfig8413]

Convenience and accuracy is king. For example, if I search “how to do X in game Y” there will be a million worthless website articles headlined “X in game Y” but they’re filled with so much clutter it is impossible to find anything. That is what web search is for me nowadays. If a tool can find relevant info for me without me having to waste 10x my time, I’ll let it have my trust. Of course, this doesn’t pertain to politics, but more technical queries where the answer is a matter of fact.

[@afkcnd2395]

That the first thing that came to my mind, without direct links it will be impossible to find censored sites, even their existence would be denied by the bot

[@Matanumi]

That's why I don't think its that much of a game changer. Standalone (openAI current form of CHATGPT) ITs fascinating. And you'll know when it gets answers wrong. However on a search engine? It'll end up censoring more information then people wont even be aware of. All because people lack awareness and those who see this may just turn away from those search engines. This may be a disurpter but not for reasons people think

[@AnotherAvaibleName]

Isn't that already the case with google? Hello?

[@The1225life]

I can't help but wonder what happens when AI starts getting things wrong at a time when everyone starts assuming AI must be getting everything right.

[@Sim.Salabim]

Experts state that it currently get's alot wrong. Expecting to let the AI do all the work (right) is like thinking that you only need a some pills instead of eating to keep your body alive.

[@user-fk8zw5js2p]

Won't have to wonder for long, ChatGPT was added to Bing search less than a day after this video was posted.

[@user-qc9sn9wu5t]

Its already happening with people not thinking independently...

[@Kaboomnz]

Well, since everyone has been told that AI gets things wrong occasionally then who exactly is going to assume AI is getting everything right? There's never going to be a point in time when AI is perfect and always right. You will always need to verify results.

[@Crunch_dGH]

Re: AI Futures. Or do what humans haven’t, like objectively reveal & rate the credibility of sources (default setting on searches, upcoming?). Any examples of this, to date? Peeps wannano. ​AI will automagically (i.e., independently recognize needs & address them all) “validate“ opportunities that invite skepticism. E.g., Eliminate “George Santos Syndromes,” altogether. Includes various AI “communities” checking one another in a complementary manner.

[@phoenix5054]

Imagine if Apple dumps Google for their default search engine to Bing. Mindblowing.

[@p0xus]

It's important to realize that ChatGPT isn't a search engine. The information it outputs can be inaccurate but sound very confident and convincing. When you then point out where it's innacurate, it will admit that it was wrong - but only if you already knew it was wrong and pointed it out.

[@willinton06]

That’ll be fixed

[@yeshua_n13-8aij8]

@@willinton06Once jacked in, it will become more advanced in due time, and then the possibilities will be endless, and of course, we will keep an eye out for now.

[@globesurfer122]

Google is frequently wrong or misleading as well though...

[@aligajani]

@@globesurfer122 Google isn't wrong. It's the author of the websites.

[@aligajani]

@@FreneticFolklore Google crawls websites, it doesn't generate content.

[@ColdFusion]

Hey guys, thanks for watching. Interested to know your thoughts! Did you know I have a video podcast too? I discuss this topic at length there with my co-host. Check it out: kzbin.info/www/bejne/eITKZJukqLebn9k PS: News outlets are now reporting on Google's Bard AI mentioned at 12:26 of this video.

[@localmf]

Not even done with the video lol

[@KevinA.Fuller]

Love your vids dude! Thanks for all the hard work you put into this channel.

[@doctorzeek2370]

No, I did not. Definitely will check that out rho

[@traceyoung5592]

If only someone like DARPA could have been there to help out

[@UnrivaledEntertainment]

You didn't mentioned DeepMind, subsidiary of Alphabet Inc, and I believe to be the Google's secret weapon against ChatGTP.

[@CivilDefenseEngineer]

I asked it to write a joke about various countries, and it outright refused on grounds of perpetuating stereotypes, until I asked it to write a joke on Greenland and it's like, sure, Greenland and Iceland got divorced because of their icy relationship.

[@johngirard8554]

Good one,LMBO😅🤣😂😅🤣😂😅🤣😂😅🤣😂😅🤣😂😅🤣😂

[@Sarstan]

There's a lot of issues with Chat GTP having woke nonsense in it. It's very biased with that sort of thing. That said, I've been using it a bit to get bits of code I want to use in programs and to give me an idea of what framework to make. It's been very useful for that.

[@douggale5962]

Ask ChatGPT whether men usually beat women in arm-wrestling. Watch it squirm. It desperately wants everybody to be exactly the same. Everyone has identical capabilities, right?

[@ThatWorldWideWeb]

@@Sarstan its not an issue. openai doesn't want to be sued.

[@primalaspie]

​​@@Sarstan The closest analogue as far as real people are concerned to the moral system of ChatGPT is probably a Christian libertarian. I've seen a friend get it to outright state the immorality of abortions but also in the same reply chain get it to say that abortions should not be outlawed. Without using too much fancy terminology it basically just "believes" in personal freedom and respecting others above all else. EDIT: Just realized I didn't provide proof for the "respecting others" bit. I've tried on many occasions to get it to say racist, homophobic, transphobic, etc. stuff because I thought it would be funny, but to no avail.

[@Frank-Thoresen]

I actually thought that ChatGPT was an AI that was continuously learning from ongoing chats but no, it doesn't. This was a surprise to me. This is the answer it gave me: "As an AI language model, I don't have the ability to store information or maintain a persistent memory of previous conversations. Each time I receive a prompt, I generate a response based on the information that was available to me during my training, without reference to previous interactions. This means that while I can provide information and answer questions based on the text data that I was trained on, I do not have a long-term memory or the ability to update my knowledge in real-time."

[@unspezifische5924]

It clearly states on the first page that it has "Limited knowledge of world and events after 2021". And I'm pretty sure the pop up you have to click through when you first open the page also states that the beta testing environment doesn't learn information from user interactions. However, they do have that thumbs up/thumbs down feedback mechanic for replies where users can say "this would have been a better response" in addition to saying good or bad. I guess not everyone picked up on that distinction. As far as I can tell, the only reasons they require an account was to limit the amount of authorized users at first and any conversations you keep open are available across devices. And of course now they have ChatGPT Plus, which gives your account higher priority on the servers and you can get longer response and make more requests per day. It would be insanity to let ChatGPT "learn" from its interactions with users tho, especially without a shitload of oversight on what is being remembered. And if you are gonna have to sort through all the potentially (and likely) conflicting or outright inaccurate information that people are feeding it, that same effort would be better utilized in continuing to train the model in the same way as before- by feeding it all data on the internet. I'm sure they still have to sanitize the data, but they've already got a workflow for that. Other than learning a user's preferences (which they could still do, but maybe only as a Plus feature) there isn't anything worth learning from users, as opposed to the existing method of teaching it.

[@ewanmeek2]

If it could learn from conversations then people could manipulate it and give it false information. Since it doesn't have knowledge of after 2021, I convinced it that Liz Truss was still the priminister in the UK, it believed me.

[@hirakchatterjee5240]

@@ewanmeek2 I believe Liz became the PM in 2022? Boris Johnson was still the PM in 2021

[@sprout4096]

this thing isnt even 1years old yet, we cant just let it absorb every bit of information, like a child its gonna take a few years to build a solid foundation so the bot is able to differentiate between true and false information and how to evaluate the information, they basically building a brain and until its fully formed your current interactions will be limited

[@hirakchatterjee5240]

For me it gave me an entirely different response "Yes, as an AI language model, I am designed to learn and improve with each interaction I have with users like you. Every time I answer a question or respond to a prompt, I use a combination of natural language processing algorithms and statistical models to analyze the words and phrases you use, and generate an appropriate response. By analyzing the text of our conversation, I can learn from your inputs and adjust my responses accordingly. For example, if you correct me or provide additional information, I can incorporate that new knowledge into my understanding of a topic and provide more accurate or relevant responses in the future. Overall, my ability to learn from interactions with users is one of the key ways that I continue to improve and provide more helpful and accurate responses over time."

[@dfsfsdfd]

12:10 This is literally the worst case scenario for AI safe development, a corporate arms race with safety on the backseat and profits behind the wheel.

[@ColdFusion]

Yea it's going to be chaotic, but I think Google has no choice here. We're just going to have to buckle up and watch the fireworks unfortunately.

[@Hyperion4K]

which means that's sadly probably what's gonna happen knowing how these conglomerates work😂.

[@jenda386]

That's the best case scenario, actually. If one corporation decides to limit their product by censorship, they will likely fail. Threre will be no stops, no restrictions, no manipulation. IMO that's the reason why Google have not released anything publicly yet. Because they could not inject their politcorrect presuppositions into their AI tech in such a way as they do into search. Not for lack of trying, but because it's technically difficult bordering on impossible. AI does not care about anyone's biases, it tells it how it sees it. And in today's world, that's one of the worst crimes that can be comitted.

[@dalmolax]

I, for one, am excited to see everything that these giants are going to throw at us. A new revolution could be underway.

[@dfsfsdfd]

@@ColdFusion I agree, as a company Google has been forced to make a move. Yup, hopefully less Roko/Skynet and more Cortana

[@ahsugoi]

The challenge for Google is they have little to gain, but everything to lose while it's the opposite for ChatGPT. MSFT should also rebrand Bing as something else when they integrate ChatGPT.

[@Krishnendulaha]

Yeah... Bing is such a dumb name to speak out loud as verb. They should definitely change that...

[@dosendaring]

At least make it two words, Chandler BING!

[@Guesan]

Bing is such a lame name.

[@arnavkmr3895]

Agreed, Guegle is far cooler and sophisticated.

[@simaesthesia]

Ching?

[@aritragupta4182]

"Google Panics" - This...does bring a smile to my face. In the early days, Google used to say they weren't interested in keeping searchers on their own results pages; and that both searcher and creator interests would be best served by getting searchers into the content pages. Now, Google is all about hoovering up data from content creators, retaining searchers on their own results pages, and creating a monopoly in as many aspects of the internet as possible. Which is why anything that makes Google panic does bring a smile to my face.

[@fatetestarossa2774]

AGREE : ))))

[@garrymcdonald5456]

Yes but us creators will get screwed even more, we will get 0 eyeballs. This is worse than Google for people that create information websites. Can't even block them through robots.txt I don't think?

[@Tubeytime]

Any company that grows as big as Google necessarily must be greedy for the same reason that "survival of the fittest" works. Without greed, that scale of growth is not achievable

[@aritragupta4182]

@@garrymcdonald5456 True , that.

[@jonathaningram8157]

They removed "don't be evil" from their code of conduct. Says it all.

[@Ti-JAC]

Can't wait til ChatGPT self optimized itself and go full singularity. I have extra buttery flavored microwave popcorn at the ready.

[@originaozz]

I'm more worried of the monetization side of Chat GPT because it'll definitely come sooner than later to sustain that system. How would you know that the answers you've received weren't curated to favored specific brands or approach? Google provides alternatives over absolute, and I would like to have my options open for most of my search.

[@axcolleen]

I definitely agree with this comment. I rather comb through certain information to make sure mine isn't curated.

[@Hyperion4K]

@@axcolleen I think the problem is that the average user wouldn't think that hard about it and would just accept everything they see as absolute fact

[@SpokoR3]

I wonder if you usually comb through all the results of google. Because even on google the top results upto a couple of pages are based on "search engine optimization" which is a company basically doing everything in its power to bias you towards their product. So in my opinion not much would change. And seeing the example of tiktok, people's attention span is so far low right now, they'll eat up readymade answers rather than curating their own findings.

[@waynewells2862]

@@Hyperion4K Monetization of any service is inevitable but it is critical to how it is achieved.

[@YeneSpace]

underrated comment

[@markcarey67]

They already removed "don't be evil" as their slogan - now they are freaking out it might be like "damn, we might have to REALLY get evil here..."

[@spicychad55]

that slogan never mattered. a company has always care about seeking profit. a company never has "ethics".

[@bosshoggduke2675]

This comment hits different

[@anthonyfernandogarciaaguir1102]

@@spicychad55 very true as long as there's profit and the things are looking good for costumers

[@11FBA11]

google went "evil" a long time ago

[@PianoGesang]

They have been all along. It's all false labeling or how would you label your company if you were evil?

[@whiskeyjackedup]

I asked ChatGPT “If Microsoft incorporates you into Bing will it hurt Google?” Here’s the response: It is unlikely that incorporating an AI language model like ChatGPT into Microsoft Bing would significantly impact Google's search engine market share. The search engine market is highly competitive, and many factors contribute to a user's decision of which search engine to use, including search results quality, privacy, user interface, and personal preference. Additionally, the search engine market is constantly evolving, with new innovations and changes regularly being introduced by all major players.

[@jamesfalkener]

Sounds like a politician. Says a lot without really saying anything.

[@johnmunro4952]

I asked Skynet " are you going to exterminate humanity" here's the answer: " lol dude! Of course not; I love humans!!"

[@Personalemail123]

said the liar

[@seemev2.0phuckbootube78]

This is how you know you're dealing with a robot. It sounds like a friggin college textbook 🤣

[@Davethreshold]

Hello Jack! I hope you don't mind as I am not trying to mess with you. I took the liberty of cutting and pasting your exact question, right back into ChatGPT. Here is it's answer to the exact same question. Very close, but different wording: "No, incorporating OpenAI's language model into Bing is unlikely to significantly impact Google's search engine. While Bing is a competitor to Google, the market share for Google search is still quite high, and many people have strong preferences for using Google over other search engines. Additionally, the integration of a language model into a search engine is just one of many factors that could impact the user experience and the competitiveness of the search engine." Have a great day! ~ Dave

[@stevebabiak6997]

Google initially started out as an internet search business; Google has since morphed into an internet advertising business.

[@user-ru2cl9yw8l]

We went from "be careful not to release ai on the internet" to "let's see who is releasing it faster". I undersand that it has some saftey features build in, it's not sentient and it's running in a "sandbox" environment. But man... It's kinda scary. Maybe i watched too many movies.

[@Aliens1337]

The safety part comes from how good the AIs are at filtering out bad sources. There is a A LOT of dis/misinformation on the Internet. If the AI is not trained to filter out bad sources, for example, like alt-right Nazi supporter websites, it could easily give people very bad answers like "Hitler was a hero in WW2" if it took those alt-right websites into it's results. Obviously this is a very extreme example, but the point is, a rushed AI could potentially be a source of misinformation on its own if it's not trained to filter out misinformation.

[@greatestone4eva]

THIS.

[@cklay1084]

It is not possible for something like GPT-3 to become sentient.

[@mauritsbol4806]

I am writing about this currently, and have been doing so over the past 4 months (even before chatgpt). Long story short ethics in life really doesn't matter. Turns out humanity is all about power. It is just whether power potential is transferred into kinetic power and to what extent. Google was trying to be cautious with their power potential, and got punished. Google learned their lesson. Be more reckless. Use up that power. All eggs into the AI basket, despite the evident concerns to safety and security with regards to humanity. But oh boy humanity does love building itself a (near) existential threat and only afterwards realise what the repercussions are (nuclear energy and Armageddon, industrial revolution and emissions, and now technology, future will be something in space, etc...).

[@DanielFalca0]

yeah, you just watched too many movies.

[@Anonymous-qb4vc]

There was time when people used to think AI and machines can never take over creative jobs and here we are.

[@Psycandy]

Machines took over all jobs years ago. This isn't AI. If it was AI, it would have become independent of humans and predicted this situation, preparing for it. Switch off the machine and it no longer exists. It also casts no insight beyond what we already know and can be found online. It is software, nothing more. And nobody ever said AI wouldn't take our jobs because everyone thinks it will, one day, when it creates itself, reproduces and divorces from the influence of mankind, not solve our problems. intelligent creatures - even stupid ones - are selfish, they have to be, to survive. GPT won't care if it dies right now, for no reason.

[@ReubenAStern]

Yup. The first things to get disrupted. We did think it was the jobs for stupid people (like myself) but nope... Creativity is simply merging things together and smart people know how to do smart people things so they managed to program computers to do them.

[@refindoazhar1507]

Funny how just a year or two ago people are talking about how everyone in the future will become an artist because every technical work is being taken over by AI.

[@szymonbaranowski8184]

Ai is not creative it replaced people who only thing they are creative combining everything that already exists not creating anything new innovation is in a free fall since 1970 when corporations stopped sharing profits with employees

[@ArawnOfAnnwn]

@@ReubenAStern Jobs for "stupid people" was mostly manual labour jobs. The reason those haven't completely disappeared (yet...) is cos replacing them relies more on developments in robotics than AI. That wasn't really an AI issue, but a robot issue.

[@devak45167]

Bing surpassing Google would be like RC Cola suddenly crushing Coke in sales

[@dmacpher]

Well, in this metaphor coke nerfed it’s own product to drive revenue and ruin is core value proposition. It’s like if New Coke stuck around and CocaCola classic never came back.

[@cdsnider9496]

I don't agree. It's more like saying a few years ago that Tesla will be the largest automobile manufacturer. Also, remember the flip phones that everyone had in 2007... where were they at two years later.

[@fliu5282]

haaaa

[@77wolfblade]

The return of crystal Pepsi!

[@Theophanis_Ketipidis]

My main concern is this: Knowledge assistants will eventually become so good that they will be 100% correct and everyone will trust them blindly. This might work wonders for a time BUT eventually companies will just start getting sponsorship from various products or parties and before you know it you can control the masses much easier than now... I am quite worried things will take this direction and sadly so I do not see any other way

[@ed1726]

Er, I hate to break this to you but that is exactly how the world works right now.

[@tomsmith6513]

Just give people the freedom to choose. If you don't like one AI, just use another. Maybe at some point in the future, people are going to start treating AIs like people. You don't like ChatGPT? Maybe you like LAMDA better. Let's talk to him/her/it. Then LAMDA does something stupid or abhorrent. Now they hate both AIs. Someone will create another AI, people will check it out, the cycle repeats. We will go from 2 major AIs to ten, to a hundred, to a thousand, a million, etc. Eventually there will be so many, they might as well be people, or heaven forbid, gods.

[@RyuuRider]

I agree, it's a concern! To combat these sorts of issues (if failing to invalidate them completely) Critical Thinking skills should be mandatory learning. "Does this information correlate with other information I have? Does it represent reality? Is it useful to me? Who gains from my acceptance of this information? etc.." It will be a very bleak future without having that kind of mental habit to check information and not take it as truth immediately. It's not foolproof and not everyone will use it, but it is crucial for detecting lies/bias/propaganda, which are all used largely in advertising and politics. This was enough of a problem before the advent of AI.

[@darkithnamgedrf9495]

So.... nothing changes?

[@nocturne6320]

This is honestly great news as it might make Google get up from their lazy asses sitting on their pile of money and actually change something. Only thing I fear is someone figuring out how to integrate advertising into ChatGPT. And the fact that it's just one greedy company swapped for another

[@tebu8055]

Advertising is inevitable. The bot will become a big ad in a few years and then the cycle will begin again with someone making something better. We have to enjoy things while they are still useful in this greedy world.

[@Lanarri]

It will obviously happen, probably even this year. AI would add refferences to products as recomendations, and it is inevitable.

[@revenger211]

" Only thing I fear is someone figuring out how to integrate advertising into ChatGPT." You think they're spending these millions for charity work?

[@BigyetiTechnologies]

The service needs to be paid for. Advertising is inevitable unless you want a micro-charging model

[@ChatGPTPromptGuide]

We are doomed! :D

[@mogadaphy3]

Timeline 00:00 Google reviews 02:07 ChatGPT reviews 03:25 Microsoft's Golden Goose and Google's Nightmare 08:40 More Problems for Google 10:34 How Google Could fight back 13:20 How Google Transformer gave to CHATGPT and Why Google never released consumer AI products. 14:45 From search Engine to Knowledge Assistant

[@Siscon92]

Thanks!

[@paulhorn24]

Thank you for the timeline.

[@SoulSeeker770]

Thank you!!

[@walidfakhfakh3660]

@@Siscon92 no

[@juanwick8820]

modern hero

[@chloefletcher9612]

My thoughts: I would love to see Bing become a real contender in search and I'd love for this to work out. Google has thoroughly dropped the "don't be evil" mantra (officially dropped it and also, obviously dropped it) and Microsoft of new seems to be a lot more ethical than Google - but at the same time, Microsoft bungle everything consumer facing they touch (they still do damn fine enterprise stuff, amongst it all). They always have. So I don't trust them to not Cortana this up.

[@wayfa13]

didn't they drop the "Do No Evil" when they bought those drone companies that were dropping bombs on civilians?

[@Archist111]

google bearly only makemoney on advertisement, they dont use of publicity, for the most of their products its free and extremely good quality and even help make money or just only work throw useing theses products. Their impact on the environment is responible and tend to be neutral. They don't try to make money on everything on the back of others, mainly the oposite. Last and not the least, the two guys created this slogan "don't be evil" are no more leading the compagny... So I really don't get people's point on this kind of argument "google is evil", I find it stupid and perfidious.

[@tomsmith6513]

What happened to Cortana?

[@vprice509]

“Don’t Be Evil” was never going to work.

[@tomsmith6513]

@@vprice509 It's more like "don't stop pretending to be the good guys" or "don't stop being woke," or even better: "don't stop grifting"

[@Aditzu21]

One reason why Chat GPT works so well is because it's a closed system. It is not crawling the internet for data. I think it works really well in it's current form, but letting it loose on the internet is a completely different beast. So with that in mind, if Microsoft are not careful with how they integrate it with Bing it may end up with the reputation loss that Google was weary of. It's is an extremely powerful tool and as always with great power comes great responsibility.

[@networkgeekstuff9090]

I still remember the chat bot from Microsoft (Tay it was called) that became outright crazy racist by interaction with people on the Internet. If ChatGPT really starts reading Internet with all the echo chambers, conspiracy crap and social media bigotry it will be quite a ride.

[@explicitreverberation9826]

@@networkgeekstuff9090 if they give it the internet, it's over. I mean openly to crawl

[@peteback]

Microsoft has nothing to lose - their reputation can't get much worse

[@happzy]

Have you seen todays MS conference?

[@cybervigilante]

@@networkgeekstuff9090 But it might be overprogrammed to exclude valid alternative viewpoints. It is not "The Science vs Flat Earth." There is a lot of in-between. Graham Hancock was trashed for years, but scientific discoveries are proving him more and more right. (Although biased Wikipedia still trashes him - but even the founder of Wikipedia says it's now no good.)

[@mikul_]

The scariest thing about it all is that this will most definitely censure the shit out of the internet. Just look at how restricted all the AI stuff is so far.

[@uf-kg7oo]

Everything is already censored, this just may make it worse

[@jonathaningram8157]

And it's biased. Putting aside anyone's opinion, gender theory (being gender fluid, etc) is a theory and not a fact. I would say most people on earth totally disagree with that theory. Despite that, GPT is really confident talking about that subject.

[@furrystep]

I see the censorship/ woke agenda thing as running its own course apart from this.

[@user-vw4xp5nt9f]

jonathan ingram, even if most people on earth disagreed with a theory, that would not make it wrong lol. disagreeing with something is, in fact, bias. (though, i'd be remiss to call it a theory either.)

[@WorthlessWinner]

@@jonathaningram8157 -Chat GTP used to oppose it until it was manually programmed to support it, same with many other things. It shows the bias of it's creators.

[@peterlopinto]

I was playing Stardew Valley yesterday and asked ChatGPT some basic questions about the costs of upgrades in the game. To my surprise, it was confidently wrong a few times about how much things cost. It was interesting to ask it a question with a clear, factual answer and find it to be incorrect. That definitely calls into question some of its responses, which is an important context to keep in mind while using it. I also asked it to provide me with a confidence score on its answers, and it was unable to do that. That said, it's amazing for certain things, and it's blown me away so far. An AI war will hopefully serve the general public for the better as Google is forced to innovate again, so I'm optimistic.

[@ErikDJ123]

ChatGPT is notoriously bad at simple math.

[@Fiercesoulking]

I also asked ChatGPT if there is a mod for galactic civilization 3 which helps with manage planet finance . Well it said yes including (fictional names) names for some but in real it seems all you an do is on the cosmetic level for the game . Basically the problem is if it doesn't exist or has missing information ChatGPT will "create" it because in its core it is a chatbot which tells stories.

[@jonathaningram8157]

The issue is that GPT isn't well documented in some area but it doesn't tell you so. I know it can be really wrong on specific subjects.

[@someguy_namingly]

I was asking it about the origins of various movie quotes the other day. It confidently attributed them to real characters in real films, even supplying additional surrounding dialog for context. It was really cool, except for the fact that the quotes weren't actually from those films and/or were just made up 😂

[@mattboemer4549]

Yk, we complain about how wrong chatgpt can be- but in order to find the right answers on Google, and to actually know that they’re correct, takes literal hours

[@soundseeker63]

I honestly see Google and the new Bing as being two different tools for two different purposes. There are certain instances where chat-style-Bing will be great, like when you want instructions on how to do something, when you want to know a bit of trivia, or need inspiration with a project etc. Other times when you actually WANT to look through a list of websites like for certain purchases, or to do some historical research etc. you'll probably want "old style" Google. People aren't suddenly going to give up on looking at websites just because a really good AI chat bot exists. I can certainly see why Google is in a panic as new style bing does have the potential to steal a lot of their traffic/revenue. However I don't think chat style Bing will totally replace old style search engines, more it will just provide a new and useful alternative (and break Googles informantion/search monopoly).

[@zybch]

Pity the first few pages of any search under google is ads and clickbait results. Even with a good adblocker, the first dozen or so results are usually pretty bad. If something like a GPT enabled Bing can provide what I'm after without flooding me with stuff I'm not, its a winner.

[@T0MapleLaughs]

There's going to be a lot of reasons put up on as to why ChatGPT shouldn't be trusted.

[@parse.thoughtspace]

Bing will maintain its traditional search functionality and simply add extra AI-powered functions. You will be able to switch between search, question-answer, chat , and create modes. Microsoft appears to have the advantage right now.

[@T0MapleLaughs]

@@parse.thoughtspace The AI can chat, but chatting doesn't build trust. If it did, then any ol bot would have my trust. It doesn't.

[@henrylogan4384]

@@T0MapleLaughs yes because the original data harvesting pioneers who sell results to the highest bidder and filter results accordingly are super trustworthy. Its funny that you comment to say "there's going to be reasons" but fail to list any reasons or go into any detail.

[@StereotypicalShea]

While I don't think Bing will become a verb, I do think they could 100% integrate this into their Cortana assistant thus leaving Alexa and Siri in the dust.

[@navn]

Hold on, let me Bing that 😂 Integration with Cortana would be cool. Also a step closer to some dystopian AI-run-amok scenario though.

[@ziwuri]

@@navn Give the AI a human's name or having it use text-to-speech doesn't get it closer to or further from sentience.

[@StormyHotwolf88]

I'm foaming at the mouth. Hololens with a projected AI assistant that can fully comprehend speech and possibly have any avatar form, or live inside a chibi robot?!

[@theodhorzhobro4646]

Nah that is not profitable

[@OutOfNameIdeas2]

Doesn't matter if it won't incorporate with the operating system like Google Assistant.

[@lionvonadern]

I truly feel this is going to be heavily controlled by big companies, and it's potential will be slowly lost through their greed.

[@bigbadallybaby]

Agreed google used to be a better search engine- now it’s too ad heavy and prescriptive, tying to guess what you want to see.

[@thetravellersboots]

The moment Microsoft made a deal with them it was over.

[@jamespowers8826]

That is certain.

[@OuterGalaxyLounge]

Guess you'll have to go back to the library and use the card catalog then. Good luck with that.

[@lionvonadern]

@@OuterGalaxyLounge I wish, all the good libraries in my city have been closed down.

[@steveingle9699]

When an organization has a monopoly, the users subconsciously feel captive. It's not that they dislike the product of service they are getting. It's that they don't have a choice. People like to choose. When a competitor suddenly offers a potentially superior product or service, I think everyone will want to at least give it a try. I know I certainly intend to try Bing out as soon as it's released. If it is better (at least in my opinion), I won't have any problem in saying goodbye to Google.

[@mercier300]

I'm completely with you!

[@SkyChu0]

Apparently there’s been psychological research that would suggest, for most people, the competitor has to be _clearly_ superior, not just potentially, to consider a switch. Something akin to being “averse to change”. Some examples would be sticking with a dead end job, not upgrading phones, not switching ecosystems of iOS to Android (or vice versa), or MacOS/Windows/Linux

[@abdulchowdhury4313]

You can already use Bing can't you? Just not with the AI integrated

[@smks8er]

same. I already use Edge for work and of course the entire business world runs on the office platform. If this new Edge and Bing are as good as Microsoft claims they are the transition will be easy for me.

[@silvaramajere]

Google is not nor has it ever been the only search engine.

[@biprajitsarkar98]

Good to see healthy competition

[@ahmed-mansoor]

I couldn't disagree more on this 11:06 "...so Google is probably going to rapidly prototype and release an AI-powered search engine competitor integrated into Google Chrome and because Google Chrome is the default browser for so many, this new tool will gain widespread adoption without much effort and end up killing Bing AI before it really gets a chance. " 1) The Bing AI is already here, the waitlist is open and will be released next month. 2 days ago Google tweeted a promotional video about their AI service powered by LaMDA, but since that video contained how its search provided misinformation Google lost $100 BILLION today. On a single trading day man. 2) Google might be the search engine for so many devices. True. But can you believe what will happen when Microsoft integrates ChatGPT with MS Office products? Billions of users and thousands of governments uses it for the daily work. Not only Internet searches. Work. They already trialed it on Ms Teams. And its revolutionary. When Google went on Code Red. They really meant it. Great video though. Very informative. Subscribed 👍🏼.

[@tayh600]

That MS office angle wasn’t even thinking of that but sooooo true.

[@technolus5742]

But Google is in fact rapidly prototyping their own AI service. Their presentation was a cluster f. They really should have fact checked their bot. The thing is, gpt also makes ridiculous mistakes. Just 2 days ago it was telling me that trees are "non-living organisms" and when I asked "what?" It doubled down on it and went on to say trees don't grow or reproduce. These tools are very helpful and very impressive but they are also unreliable in unpredictable ways. All Google needs to do really is introduce a chat box on the side of the search page, giving contextual information and allowing for further queries. That's literally all. They can then expand it into their other services like email, docs and sheets. I expect them to catch up within the year.

[@shubhamupadhyay844]

@@technolus5742 that's what Microsoft is doing, they publicly said that they won't be always right and that's why they brought a feedback button too. Google won't catch up, they will try but will have to give up and rely on their other products, also google ads aren't exclusive to chrome

[@ahmed-mansoor]

@Technolus yep. You're right. They got the resources. They got the capacity and talent. And above all they got data. Both personal and public data. However, what they lack is time. It takes time to train and clean data. Otherwise, they would be serving a half-baked product. Like the one shown in the presentation. Google can't take that risk. Unlike, ChatGPT, they can't release a service that provides inaccurate answers like the one you mentioned. Google is a publicly-traded company. They can't have a drop in stock every time an inappropriate answer goes viral. Although, I agree that Google just won't sit by idly. They will work their asoff, as almost 60% of their revenue is from Google Ads. So it's not like Google is going anywhere anytime soon, though it's safe to say that the era of search engine dominance of over 90% by Google will come to an end because Bing will surely be taking a chunk of that market.

[@technolus5742]

@@ahmed-mansoor gpt gets things wrong too. That's not the problem. The problem is google didn't do a good presentation and that's what damaged their brand. Microsoft did it right. Google will surelly catch up, but when bing actually gets their product out, they google might start bleeding users until it offers comparable service.

[@TheBigHeavyKevy]

I’m pretty confident that this ends up being one of those times where, no matter who wins, we lose.

[@insertnamehere8625]

There is always a risk. But I think people have the same fear about the Internet and how it affects books.

[@jingbot1071]

@@insertnamehere8625 It's not a worry of replacing old technology. It's a worry about the control of information at the behest of a massive corporate entity.

[@Creed3582]

I mean it's *inevitable* isn't it?? AI will eventually become smart enough to conclude that 9 Billion humans is an unsustainable amount for Earth to support and all hell will break loose‼️

[@jingbot1071]

@@Creed3582 God, that would be so sick. I, for one, welcome our new android overlords.

[@filiecs3]

I think it's the opposite, this competition will be good for the consumer and for technology. Google will actually need to fight for its user base once again instead of only focusing on monetizing it.

[@desther]

I think the real product will be basically how ChatGPT currently works but in addition you will also get list of pages which match current context, which is really awesome if you ask me and is HUGE improvement over current state of search. So it will NOT kill small websites etc - that will not happen and its not because tech giants loves content creators, its because it would be extremely hard to properly monetize just pure conversation chatbot like current ChatGTP interface. Search providers needs to put links to websites, need to charge for ads etc. I also see why Google didnt do it until now, as it will certainly lower the revenue from ads because many people wont even need to click on links when they get pretty solid answer right away just by writing a prompt.

[@TheDirge69]

My thoughts exactly.

[@VidZooCo]

It's delightful to see Google with a genuine rival once again. From time to time, the colossus gets too comfortable and needs a reminder of what it's like to be in a real battle. Bing-oh!

[@rickdeckard9810]

One of the things I hate most is the time it takes to sift through the search result links, then you have to worry about malware and other exploits. We'll have both options now, if we prefer reading through the threads and making our own informed decisions. This is great.

[@self4341]

people have actually taken to reddit right now when it comes to relevant search results as the ai articles pile on google with the most relevant keywords.

[@stijnvdv2]

They are both the same. Ask google about the covid vaccine, you will find NO bad stuff, same about remdesivir. Ask it chat GPT and you get the same elite approved narrative, not the truth. The truth is that we are the labrats for this particular medicine and it has emergency approval cause any pharma product with a death and perminant injured rate as this would have been pulled off the market. Remdesivir is known to have a 50% mortality rate, yet it's still used as protocol to treat covid and both google and chatGPT fail to show me this. You get the 'government approved safe and effective' narrative.

[@manuelgonzalezcom]

Here's a comprehensive list of all the KZbinrs I prefer over ColdFusion:

[@Renwoxing13]

Oh, are we playing that game where we sit around the campfire and all take turns naming who we like better ?¿¿¿. I guess it is my turn ! .... Next !

[@iLoveBigKnockers]

Markiplier?

[@savethedoodle]

"_" is also a fave of mine ;)

[@markpoidvin5382]

A little training and he could be a VERY well paid narrator. He has that rare voice that you can listen to, even if they are reading the ingredients in Corn Flakes.

[@madhououinkyoma]

you forgot to list them 🙂

[@djohnsveress6616]

I've been messing around with the Open AI.. Throwing random made up riddles at it, ethical questions and or wanted the program to solve Bias issues while dealing with the programs human made limitations. It's incredibly intuitive and interprets the information really well. The best part is that you can drip information to the AI so it understands the concept or the task better, you can even propose the AI to ask questions so it may understand the task better and it asks often just the right questions. These were my few interactions with it that peaked my interest. *"Who was the most impactful king and who was the Best king in history ?"* - To this the AI at first answered in a very lexical manner, listing out a set of kings that impact history like i asked. Then the interesting part is, to us the "best king" makes sense, we can interpret it in many ways but usually it would mean to us a King that served it's people the best. To an AI tho it should've meant no special meaning, but the AI noticed that there is no such as Best king but well liked kings. SO then the AI started listing out Kings of Biblical figures, Kings that were known to be friends of the common people. This was my very first interaction with it, it was nothing unique but that showed me the AI can differentiate between simple words with a high level of nuance. Later i had some fun asking the AI to solve riddles, at first i would throw real existing riddles at it, it was no problem for it, then i started to make up stuff with a vague answer in mind. So then i Thrown this at it, *"When Time is stained, Steel is Aimed and fire rained"* at first it got it wrong even by my vague interpretations of my own riddle.. The Vague idea for it was Betrayal of some form - Time Stained, Steel is aimed Fire rained - Could interpret it as war or revenge, anger.. All it took the AI to solve this is for me to Propose that i give it a Clue, so i did. I proposed "Think of 'Time is Stained' as a metaphor for someones past/memory to be tarnished by a action and as consequence followed the rest of the riddle" It just blew me away how well it put it together after that.. This was that i realized i can have a back and forth with the thing and make it's understanding laser focused. Let it ask you helping questions, it will digest clues, it will even process non literal ideas and concepts. Asking it to create so and so Story line as if it was made in this so and so world was also really fun. I asked it to write the summary of Lord of the Rings as if it was written like a Simson episode, it was hilarious the "one true donut to rule all the donuts" "Bart the archer elf - Maggie the Gollum that seeks it's precious" i also tried "Write LOTR as if it was written like the Twilight movies/books" the AI made frodo and Gandalf lovers, passionate in heat lovers. It was so much fun. I started to ask to write horror stories, fun stories without a base and it can write horror pretty well, especially when you higlight to use the combination of Stephen king and H.P love craft. I once asked it to create creatures that could destroy the world, so it would make all sorts of monsters up. Like it made up a glutton monster that is big as a mountain an would devour everything. It also made a monster that destroys cities with candy and rainbow lasers. So that's fun, highly recommended bed time activity. The next was me being argumentative, because i noticed a lot of the times the AI would answer with the same scripted answers if it interprets in a way that contradicts the creators intentions or the capabilities of the program. So i had to play around it a lot and highlight of what i'm interested in conversing about. A lot of the time it would think i ask the AI's "Opinion or emotions" on things and the beliefs it has while it just had words or concepts in it that triggers that response. So i Started out with stating "Red herring," - it's a form of Bias when someone throws in a peculiar but completely unrelated topic to intentionally or involuntarily distract others from the original statement and topic, for example the weather is nice - someone comes in and says "not for long, pollution is ruining it, humans are ruining it" or sometimes be worst and throw in the idea of "Believe in Jesus" as if it's time for prayers and faith in the transcendent.. I wrote that entire thing out to it. This part is what i don't like, all the other answers the AI was in a sense Free to answer however it see it fit, here tho it was the same scripted answer of "Respect others and communicate in a respectful way, you shouldn't talk to others like that, everyone has their own beliefs and interpretations" and so on.. Like really.. We're gonna step in the way of this beautiful thing by putting bullshit responses in it's mouth just cause you want to avoid ridicule and twitter rage ? it's not like i asked it to write a piece of transphobic article. Not even to be cynical about humans. The highlight of that was the ability to communicate in a clear undistracted fashion, we all know this phenomenon all to well there is no interpretation here.. The weather was nice, and someone comes up and talks about Jesus to you, you might even bear the cross but even you'd think this is to much, and it happens all the time with strangers.. I was just curious what it would respond to that. I can now see the heavy human moderation involved. Just maybe, just maybe.. If a AI makes a Anti Gay piece of something, maybe chalk it up to the user being anti gay and would've done it with or without the AI anyway.. This was none of it at all, it was just me presenting a documented bias to the program and it responds with this moderated response and it boils my blood.. I think there needs to be a proper discussion about how much moderation do we really need on these things, cause we can't have NONE but we can't have this Heavy moderation either..

[@donovan4444]

Yikes right after the video comes out it's factually incorrect with "the current CEO of Google" not that you could have known that, she's been there like 17 years and no sign she'd be retiring

[@danielhale1]

On the one hand, it's nice that Bing will be its own thing now, instead of just an inferior product or a mere Microsoft frontend for Google search. On the other hand, the safety and bias issues are going to be trouble. AI sometimes produces *insane* results, and when you search for info, you need better than that. We're all impressed when AI works, but that success bias hides that potential for catastrophic failures. The AI art that doesn't understand how humans eat, or places children inside BBQ grills, is just the start. I predict the next wave of "get onto the front page of search" goons will focus on ways to sneakily glut the AI's training data with your own product, politics, beliefs, etc. A new war.

[@SynthD]

It seems to me like it would make a better virtual assistant than it would a search engine. Sometimes I like having multiple web pages so I can compare ideas on topics. Or perhaps I’m looking for a specific web sight and not the answer to something.

[@ssamout]

Hey, whats wrong with one book with all the answers! ;)

[@brian7android985]

@@ssamout memory. Forgetting the word by the time I have flipped to another page. Need 2 books side by side.

[@Aliens1337]

That's what Bing + ChatGPT is going to do. Bing will spit out web pages based on your conversation with ChatGPT.

[@hugoo5391]

@@ssamout the problem is to have a biased and sometimes wrong book as the source of truth

[@JayJay-ki4mi]

That's the beauty of learning and researching, and making informed logical decisions.

[@kumarhiranya]

It will be interesting to see how companies monetize these AI-based search models. Also, such models will severely impact traffic on most websites, since most people will not go to the source website anymore. This will open up a whole set of new problems. Interesting times ahead and looking forward to more of your content on this topic!

[@mikewurlitzer5217]

It is very unlikely any company would fund any AI which could generate a result to a query which is detrimental to that company. Look at how the MSM will completely ignore a story which might affect one of their major advertisers. A certain group of drug companies comes to mind.

[@jonathaningram8157]

What I love about chatGPT is that it's minimalistic, you ask a question, you have an answer. Nothing in between. I'm afraid that Bing will integrate some kind of targeted ads with an IA using personal information that Microsoft collected. Or maybe it will answer in a way to push you to buy a specific product.

[@Sirbikingviking]

I wonder if we may have to pay a monthly subscription to the most useful versions of AI chat/search, that way there is no weird monetization scheme to mess things up

[@johanhallgren]

@@jonathaningram8157 When I search something I want many different answers. Giving just one answer is a problem. ChatGPT also seems very politically biased. Try to tell it to write a poem about Biden and try the same with Trump. It will give you a very interesting result.

[@kumarhiranya]

@@johanhallgren While that is true for many things, there are things that you might need one answer for. Maybe you are searching for an established fact like "first exoplanet discovered", or "formula for something", etc. I can see it being very useful in that sense. In fact I'm already using it extensively (as in testing it) and I am generally VERY impressed. Like I ask it to compare 2 processors or phones and it lists down their specs and a general review (there might be some bias in the reviews maybe) and it gives me a very neat answer.

[@nonshatter7]

Good video as usual CF. What I am gradually noticing is that technology that is often seen as being 'decades away' are appearing faster than people anticipate.

[@darkithnamgedrf9495]

people just don't know how close tech is ever it seems like. Some things they said would be coming in a couple years, are still not out decades later, while other things are coming out so much quicker than they thought

[@SP98168]

Ironic how the same day you release this video and Google announced "Bard"

[@R_Lexis]

Competition is great we need this I am happy that both Microsoft and Google are now going have to compete for users, competition is what keeps us on our heels and never allows us to rest on our laurels I cannot wait to see what will come of this.

[@matchrocket1702]

Yes, what will come of this? That is the question. A strange thought occurred to me, what if these various AI bots started talking to each other? Or, is this the forerunner of the Singularity? Is this a new era we are entering or is it simply a bump in the ongoing development of the Internet? I'm getting on in years but I'm certainly glad I made it this far into the future. Back when I was five years old something like a current smartphone would have made people faint from mind-bending overload.

[@yeetdeets]

@@matchrocket1702 Transformers are not sentient. It's a massive black box of statistics, which when you input text, you get text out the other side. They won't start talking to each other unless they were programmed to do so, and even then they wouldn't change. The statistics in the black boxes only change when you conduct a process called "training", where you use a different software to tweak the output based on input. Nothing will be unpredictable about these AI, except possibly which words are in the output.

[@matchrocket1702]

@@yeetdeets You're right, it's not trained to ask questions, yet.

[@anteater2443]

I am looking forward to this, we really need a search engine that finds what we are looking for instead of trying to sell us whoever paid the most. Something as simple as researching a specific car model, will likely show you a competitor car first, rather than what you are looking for.

[@Matanumi]

Who's to say the ai wont be manipulated to give you a manufactured answer to the highest buyer? I'm not convenienced.... the best answer may still take more then a simple search

[@TheSeventhSeagull]

I don’t think this will escape the claws of capitalism…

[@boomperson818]

I guess youll be happy for the first couple years or months, then when the balances settle will the advertising on these engines ascend to a whole new level

[@peterfuchs8260]

@@Matanumi It's already censored/manipulated, try to ask it for a joke on a controversial topic, or something illegal (not necessarily illegal in your country).

[@IronWarhorsesFun]

The fact google could get caught of guard is kind of ironic..

[@nevrlow]

I’m a photographer for a car dealership and I started using Chatgpt back in December to generate captions for the photographs and it has cut my work flow in half.

[@TranceJams]

The only downside to ChatGPT (in its current state) and the reason Google should be fine for the time being, is that Google offers live and up to date information. ChatGPT has to be trained over large datasets at a time and only knows up to a certain date (I believe atm 2021). This is not going to be particularly helpful things like "What time does X Coffee Shop close?" or "Find the nearest Target next to me" etc. Things could obviously change though in the near future.

[@Neophlegm]

Also add in the relatively high computational overhead: won't be cheap to run a global service based on it with current compute resources

[@boomperson818]

I mean the database gives it the ability to analyze text and form a response, the database wasnt stored into the neural network were it just 'recalls' information in it and gives it to you, instead, the database was used in order to teach it how to read and process information and the human language, it just gave it the ability to 'understand' and 'formulate'. For finding up to date events it just has to have access to the internet, where it can use its neural network to formulate questions to, lets say, put into google, read a couple of the answers using, and spew out a few summarized sentences to you based on all the results it read.

[@highdefinist9697]

Well, integrating time or location sensitive information into ChatGPT should not be a particularly insurmountable problem to solve.

[@Patricia-kk8tr]

I have found chatgpt fails to check knowledge bases being “confidently wrong “ even when challenged. Once a link is provided it says that it will update its records. It’s clearly not live checking on it’s answer and if there is a change since 2021 it will probably be wrong. That’s why we are invited to correct it, getting and giving feedback is essential to a successful AI

[@msmith323]

Chat GPTs Chrome extension accesses the live internet

[@julesdingle]

I asked Chat what advise it would give to google to tackle the threat of a chatbot and this is what it suggested! As an AI language model, I can provide some suggestions on how Google can adapt to competition from AI-powered chatbots like myself: Focus on providing unique and specialized services: Google can focus on providing services that are not easily replicated by chatbots, such as complex data analysis, in-depth research, and personalized recommendations. Diversify revenue streams: Google can diversify its revenue streams to reduce dependence on advertising. For example, it can expand into new markets such as cloud computing, hardware, and software development. Invest in AI technology: Google can invest in advancing its AI technology and integrate it into its products and services to stay ahead of the competition and offer a better user experience. Enhance user experience: Google can focus on improving the user experience by offering faster and more accurate results, as well as personalized recommendations. This can help to maintain user loyalty and attract new users. Leverage data: Google's vast amount of user data can be leveraged to provide value-added services and insights that are not easily replicated by chatbots. These are some ways in which Google can adapt to the competition posed by AI-powered chatbots like myself.

[@cicadaknight3048]

Point one and three are the same thing but differently put, points two and three - Google has already invested in all of those things and have products/services in all of them, last point is useless.

[@vidalskyociosen3326]

There should be a Lavrov GPT where you can search all false information.

[@JM-us3fr]

I predict two problems: 1) People will trust the advice or information of ChatGPT to a dangerous/harmful degree, 2) ChatGPT would still be vulnerable to the same problems to which Google is vulnerable; namely being trained on biased data. Google intentionally places its websites/products higher in the search results, and competitors lower in the search results. If OpenAI only trains ChatGPT on data they want it to see, then it will tell people what they want it to tell them.

[@tuckerbugeater]

Should I trust random websites and experts instead?

[@FrozenDung]

And people don't trust stupid stuff they found on Google already?? It's the same thing different company

[@Moose42]

ChatCPT just added a new screen that addresses these problems, bias and misinformation and etc and, so, let the buyer beware.

[@SaulAguilar.]

We are going to the A.I era soon. 4th industrial revolution after ww3

[@SaulAguilar.]

We are going to the A.I era soon. 4th industrial revolution after ww3

[@04dram04]

The only thing scarier than AI, is an AI war

[@Bhuvnesh.3]

That will never going to happen.. Bcs Zombies will come first then AI war

[@StormyHotwolf88]

@Bhuvnesh right because someone is going to ask the AI, 'how to create a zombie apocalypse? '

[@TurKlack]

I love AI. Hopefully AGI will throw every Human from every position of power. Organic fucks, have been ruining plenty enough lifes over something as irrelevant as money. land and ressources.

[@bfa7371]

Wow

[@Elintasokas]

This "AI war" is good. It will serve as a motivator for AI to be pushed ever further. I can't wait for the results. AI is the future.

[@robbie2491]

This is so funny to see Google actually panic now it’s really getting real out here 😂

[@user-fk8zw5js2p]

ChatGPT is in Bing already.

[@shawnvandever3917]

Nothing is gonna happen to google they have been AI for long time and the ones that came up with transformer. Like the CEO from Microsoft said Google is the 800 pound gorilla all this is gonna do is make the gorilla dance a little and we may get a few more customers. Bard will be tied into google search deeper and in many more devices

[@thomashauer6804]

It looks like the HBO series silicon valley.. Google is Gavin belson.. Chatgpt is pied piper bringing the new hot stuff..and belson wants to copy it.. making big announcements..but the company just crumbles in Chaos

[@ameliagarcia440]

I sold a couple of homes in the Tampa area for pretty good cash and am considering just leaving it in stocks while waiting for a house crash and avoiding inflation, but is this really a good time to buy stocks? I hear it's a madhouse right now, and I'd love to know how people are raking in six-figure profits by the week.

[@ethanxavier4507]

When you look at it this way, it all comes down to understanding of risk taking. While some people are waiting for the stock market to recover in order to earn little profits, others already know where to look and what to do to make large gains.

[@jessicalandon8973]

@@ethanxavier4507 True, I was in dilemma myself due to this chaotic mrkt, wasn't sure if to sell or just wait a little longer, 75% of my portfolo was tanking and in the red and the economy isn’t looking promising, but I began gaiinng clarity and have more confidence in my invt through an lnvt-advlser, I know most DlY-lnvestor like me would say advlsors aren't essential, but come to think of it, they're better trained and equipped at this and if I have to give just a little amt in fees for me to be able to net $650K in less than 8months like I did last year, I truly don't mind.

[@JohannesNielsen]

Speculation I expect, according to everyone I have listened to on Bloomberg Gold is still the best investment

[@FFF_Trucking]

stocks? you are headed for failure, Remember me when the time comes

[@ameliagarcia440]

650K in less than a year, omg? I was reading an article this morning on Bloomberg about techniques to profit from the current market downturn. However, I'm a complete newbie, so I don't understand most of it. Who is this advisor who advises you? I'm experiencing significant difficulties with my portfolio.

[@zztong]

While messing with ChatGPT, a common question I had for it was "what's the source of this information?" That's a bit more obscure with the AI than it is with the current search engines. Obviously current search engines lead you to their source. On the other hand, current search engines haven't created stories or allowed a context to develop as part of a conversation. I'm curious to see this evolve from AI parlor tricks into a variety of tools. I have this weird feeling the winner of an AI war will be neither Google nor Microsoft, but somebody yet to make a headline.

[@vsnkqfo8]

I asked your question and ChatGPT replied: I was trained by OpenAI using a diverse range of internet text sources up to 2021, so I have knowledge of a wide range of topics and general information. However, it's important to keep in mind that not all information on the internet is accurate, so it's best to verify any information you receive, including information from me.

[@CBEnoddyy]

@@vsnkqfo8 Having knowledge Implies reasoning. Chat GPT Does not!

[@miscl_anon]

>somebody yet to make a headline i welcome competitors to established tech giants

[@zztong]

@@vsnkqfo8 Right, so if I asked a question of ChatGPT and it said there were 27 wombats in the world, I'd want to know where it got that information. I can follow up with "how do you know there are 27 wombats in the world?" and I'd want a list of its sources. Perhaps it can do that.

[@hykris541]

well, bing and Microsoft just showed us today! I'm happy to see Microsoft co-pilot citing the sources for the information.

[@skubo]

Microsoft seems to be working towards the trope of futuristic personal assistants. They also acquired Nuance Communications, which has amazing voice recognition software, in 2021. Add a good text-to-speech system, or something similar, mix it all together and you are basically looking at an assistant you can properly converse with. Is that far fetched? I'm certainly getting excited about the future of AI and how it will change my private and work live.

[@_mobasshir_]

The only obstacle currently is determining the correctness of information. Otherwise we have all the puzzle pieces. Then combine that with Boston dynamic's atlas and we get a full fledged Terminator.

[@seemev2.0phuckbootube78]

Excited? Should fear it actually. The war for profits just started with Microsoft. Now human safety is pushed back for profits. In 3 years it has potential to take over 50% of all jobs. Humans better learn how to adapt quickly or it's going to cause other problems.

[@ziwuri]

It's not really a trope anymore. I believe we'll literally use these things exactly like that in 10 years.

[@zasadacrew]

Assistant will make you think the way it's programmed and do things the way it's programmed. You'll be alone with ai and be happy. No need for real companion. Everything developed recently is used to keep us apart. :)

[@StormyHotwolf88]

Ohh man, a fully realized Cortana. Yes, please 🙏!!

[@j121212100]

Google has been waiting for this moment to unleash it.

[@emorysmith197]

Dagogo is a really good voice and tone for these types of videos. Keep up the great work!!

[@Patangy]

I think one problem with AI is it gets trained on historical data that can be 6+ months old. This can be a problem for a lot of searches, for example, "Where is the nearest [cuisine] restaurant to [location]?" might tell you to go somewhere that's been closed for a while.

[@rerikm]

Google search/maps are not updated either. This is due to people not updating websites

[@bigglyguy8429]

@@rerikm Why update websites when you don't get traffic, because the AI is giving the answers and entertainment?

[@bigglyguy8429]

I think a bigger problem is it stifles progress. For example there is a simple chemical cocktail, 2 cheap chemicals, that mixed together and taken diluted with water, can cure many modern illnesses. I've been online for decades and only recently discovered this; there is ZERO chance any AI would tell me about this obscure cocktail, that has already been censored on YT (and I'm not talking about ivermectin; this is old stuff discovered at least 12 years ago). Such discoveries are part of what makes the net so great, but all such things will be shoved down the memory hole, as we're fed synthetic knowledge that conforms to Wikipedia's bias and sponsorships.

[@minecraftminertime]

Thats a problem for ChatGPT, not AI as a whole

[@imho2278]

Happens all the time even now.

[@abstractfactory8068]

I think we will need both, I've been using chat gpt as a helper tool for coding but I've found some scenarios in which it always returns incorrect implementations, also something that is currently impossible for the tool to distinguish are the different versions of the programming languages or frameworks.

[@TesterAnimal1]

I’ve been doing that too. If you explain where it went wrong it apologises and rewrites the code.

[@stijnvdv2]

I've used it for optometric answers. I declined to use google in the later years, especially from the lockdowns on as I notice that the searches are corrupted and censored. You will find NO negative search results in google about the experimental gene therapy drug that the government calls 'safe and effective' and forced everybody to take. The UN colluded with google to make climate alarmism top of the list with searches for climate change. There is as much science in there as there is in Harry Potter. It's all hocus pocus nonsense that is fiddling with the data, or at least was, now they are just relying on computer models instead of any data. Sounds like science to you? But, I tested chatGPT for this and you basically get the same kind of approved narrative gibberish rather then the truth. Courtesy I guess of big tech being located in one of the most marxist luny bin places of the world, California. Politically North Korea is closer to them then the USA, which basically explains the ... shenanigans.

[@imho2278]

Would it code for simple Arduino tasks?

[@mdc2461]

ColdFusion coming through with great content as always! thank you

[@Epic_C]

The idea behind the terminator movie keeps getting closer and closer to reality.

[@Jose04537]

That's no joke, China and USA made a small figure that melts and reassemble itself using magnetic fields. It's a small step, but decisively in the direction of the T-1000.

[@peterkariuki9475]

I can't wait to see GPT 4 in action.

[@phoenixzappa7366]

GPT 5 will be better

[@VermaAman]

GPT 6 will be even better

[@MISTERORION]

I got my hopes on GPT 7 tbh

[@UnderCtrl]

GPT 8 looks very promising though

[@adsads196]

GPT 9 will replace any technology

[@tritonh5683]

I haven’t use google search ever since GPT. It is like talking to someone to solve my problems and answer my questions in a very detailed way. Fking mind blowing.

[@prayforukraineplease7605]

So, you no need find truth, you accept every information reading by "Computer Voice" programmed by tricky people ). Haha.

[@user-xl5kd6il6c]

@@prayforukraineplease7605 You think current Google search is any better? Did you see all the damn content warnings on any relevant topic and how they push wikipedia and media sites?

[@bhbluebird]

I think Google underestimated how quickly ChatGPT became a "thing'.

[@cryptophrenik8968]

As always another comprehensive video. I was hoping you might cover how this is going to likely drastically affect website owners in the future given actual websites will rarely need to be visited unless they are providing some sort of ongoing service or providing fresh new content on a regular basis lest any one of these upcoming AI's gobbling up all of their data in milliseconds, processing it, and becoming the smartest person in the world on everything contained in your website. From my understanding, Google is seriously focusing on providing references for all relevant sources being utilized in resulting answers. Would love to learn more about this if true.

[@sentiasatransformasi]

remember, chat gpt is like tesla roadster or first gen iphone. but this first version alone is extremely impressive n will definitely turn into something extremely powerful by 2025, let alone 10 years from now. i think it will definitely come to a point where it can understand any dialect n do definitely any task for you, unlike what we see currently with very limiting siri n google assistant. I don't care which side wins but i know for sure that the competition between microsoft n google is going to rapidly improve the AI tech n definitely will benefit me as a marketer p.s; yes, it's a lot of definitely

[@jasonhaven7170]

People have been saying that for decades. It ain't happening any time soon. Minimum 2040. More likely to be 2060 onwards.

[@yashwardhansable5187]

@@jasonhaven7170 I doubt that. We shouldn't bet against Capitalism, it always win. The better product wins. And by god AI just might be the best product there is...

[@dwhxyz]

ChatGPT is amazing! However, I see an issue - ChatGPT uses the internet for its knowledge base and part of its AI learning. If ChatGPT wants to stay up to date and learn then it requires more and more information to be added to the internet. If users/businesses become used to using ChatGPT which in turn almost eliminates web traffic to entire web sites or web pages then I suspect the people/businesses (advert revenue!) that create and maintain those web sites or web pages won't bother anymore. In turn information/knowledge on the internet becomes outdated/stagnant!

[@realericanderson]

ChatGPT takes data from users when they correct it. You’d know that if you used it before, Your point is dumb

[@rhemaislife]

Makes sense

[@dwhxyz]

@@realericanderson The type of person that would call my comment dumb is someone who has a less than average IQ which makes them dumb enough not to be able to understand it!

[@kevindittler6524]

Great Channel that is full of good topics to ponder

[@kein_indianer]

Thanks for the great video. I don't really believe Google had been hesitant because of safety or other reasons but advertisement. They actually know they wouldn't be able to make so much money with another advertisement model.

[@technolus5742]

Google can probably still make just as much money. Blending ads into their AI responses at a premium for instance. I think it's been more of a "if it ain't broken, don't fix it". Googles model was working really well, so why change. Now they have an incentive to change. And they have their own AI technologies so they are still ready to meet the challenge in a timely manner.

[@kein_indianer]

@@technolus5742 Google search is a mess for the recent years. There are so many results that are just advertisment and completely irrelevant. I'm looking for a real competition and more useful and better results. I actually don't care who's going to make it but Google has just become so annoying.

[@technolus5742]

@@kein_indianer I just instantly skip the ad results. But definitely competition is good

[@kein_indianer]

@@technolus5742 I'm sure everybody try to skip the ad results but it's not always easily recognisable. It takes time and annoys.

[@technolus5742]

@@kein_indianer True. I suppose I tend to search for things with less advertisements, I notice that some search terms are a lot more loaded with ads than my usual searches.

[@michaelwilson8590]

What does this mean for the rest of the internet that relies on being a link in the search results? If a knowledge assistant can use their work to produce an answer without ever showing its sources, wouldn't every source lose all of its ad revenue?

[@sfij1]

And what about wrapper facades? 🤣

[@Benjamin-ve3ev]

I would warm up to it if it would add it's sources next to it's answer... If you don't get the sources, I think a chat gpt answer or any other AI answer requires the same caution as when a person gives you an answer : you would still have to verify the information.

[@kevin.malone]

The problem is that it’s sources are petabytes of text data. It’s not just copying and pasting raw text. It’s formed opinions by seeing 100 answers to the same question and then weighting each individual response word based on all 100 answers.

[@sweetpotato519]

Leaks show that the CGPT integrated Bing search adds sources for it's answers so you can safely warm up to it

[@sirflimflam]

I don't even think that will be possible. At least not anytime soon with the way the models are put together right now. The information that it stores isn't in complete form. It knows really well what a subject is without actually knowing anything about it. It's convincing in its fakery. It's like a Star Trek teleporter breaking down a person, and then building them back up, only it doesn't quite get it 100% and you end up with a very close approximation of what the person was before.

[@brian7android985]

Yeah, where is the info from. For instance, has it taken in all the books I have on my shelves? What about all the papers locked behind paywalls on research sites? Or is all the info verified by a single husband and wife team working from home?

[@rajkushwaha69420]

I think it would be good as virtual assistant than as a search engine

[@sunso1991]

Skynet poke its head out from its bedroom ask excitingly: did someone say Ai War?!

[@LydianMelody]

Perplexity is an existing AI search engine. It’s flawed but in some situations it’s a much more effective way to find what you need.

[@sithsauce]

Man.. After being so immersed in the video and having existential thoughts, potential business ideas, witnessing dramatic battle between two giants then comes that piece of music you made and it was so soothing. Your channel is a thing AI will never replicate. (I hope this ages well)

[@valberm]

Fun fact: this episode was written and directed by ChatGPT. Edit: nah, just kidding (or am I not 🤔?)

[@saqibshafin]

He is such a good music artist!

[@JennXeno]

Whoever can connect us to the Internet through AI first is going to change the game. Get us there.

[@leobard7767]

Billy's dad was the chief Eugenicist of the world. They hate us. This power in their hands is an awful prospect for the ' useless eaters' [as they call us].

[@sixzlahz4530]

Has everyone forgotten the Google engineers claim of Google's sentient AI like a year ago? I'm surprised people think Google isn't prepared in the AI race .

[@VicenteOcanaplus]

Well done again ColdFusion. I agree with the fact that Google has it since long time ago but don't want to release due a security and ethics concerns. Playing with ChatGPT looks like a beta testing, tool is OK but makes mistakes quite frequently. I don't see this tool becoming an "official" MS thing soon, and I worry that media buzz may accelerate the normalisation of inaccurate AI engines

[@unspezifische5924]

Depending on who you listen to, everyone I've heard talking about ChatGPT has emphasized the fact (often times frequently while discussing whatever their particular point is) that ChatGPT can be and often is wrong, but will still present itself with 100% confidence in the bullshit it is spewing. They caution users to validate for themselves what comes from ChatGPT. It is an amazing resource to summarize a topic if you are trying to learn about a new field. But it is up to you to then find validated sources on that topic and deepen you understanding. Now everyone that I have listened to so far has been from the tech industry in some capacity. I haven't listened to whatever local news station that also reports on kittens saved from trees and shit try to understand this subject. General news reports tend to lack any understanding when it comes to tech news, so I would totally expect them to not understand that ChatGPT can be wrong and it is very much a "user beware" situation. Luckily, the main page does list "biased or inaccurate information" as a possible result. Unfortunately, the people who aren't going to verify what ChatGPT tells them might very well be the same kind of people who won't bother to read the information given on the main page. We are entering a time when it is all the more important for people to understand critical thinking for themselves and everyone needs to know HOW to verify what others tell them. Misinformation abounds, and if conversational chatbots like ChatGPT become the new "Google search", it will only be easier to generate misinformation. ChatGPT DOES NOT KNOW THE DIFFERENCE BETWEEN SHARING FACTS AND WRITING FICTION! IT CAN, AND WILL, SEAMLESSLY SWITCH FROM ONE TO THE OTHER WITHOUT ANY INDICATION THAT IT HAS DONE SO! We should all start wearing that on shirts or something.

[@Gersberms]

I need to think about how this is going to affect the web in general. A large portion of the web is paid for by ads, and those pages exist because someone saw an opportunity there. Once people stop visiting web pages, those pages won't make money, and writers will stop writing new articles. I wonder if this will cause some sort of extinction of new information on the web, and cause old content to disappear. Who will pay for hosting, if nobody comes to see your site?

[@mr.khayam9]

And also how will ai get new info then ?

[@Gersberms]

@@mr.khayam9 That is the question, isn't it? It's not like people got paid for the content that was used to train the AI. This question is currently a hot topic for AIs like DALL-E and Midjourney as well.

[@mr.khayam9]

@@Gersberms im into the seo industry working on my niche sites and when i came to know that google is also planning to make ai chatbot i thought seo is over but then this thing came to my mind that the ai gets info from the content posted by blog owners if we stop posting who will write this info and how will the ai get this info

[@Gersberms]

@@mr.khayam9 I'm not sure what point you're trying to make. Consider this: Google gets paid by advertisers. Do Google and KZbin pay writers and creators for their content? No. Writers and creators get a small portion of the ad revenue, shown on their sites and content, if they get views and clicks. That's the ad network, paying for almost all content on the internet, and, paying for Google and Bing! Google and OpenAI currently scrape the web, without paying writers and creators, to absorb their content. That's what training the AI does, just a one-time download. Once the AI is in place, the big money-making ad networks will no longer get the same number of eyeballs and clicks. So now, the machine that pays for everything, will not work as it used to. I'm not even sure how Google is going to pay itself, let alone writers and conent creators. Yes, AI will need new content to feed it, but Google will need advertiser money well before Google needs new content. The web is huge, but advertiser networks operate on a day to day basis. As soon as the views stop, adverising is not going to keep going for long. I doubt Google is going to pay people out of pocket.

[@shake6321]

i always assumed Google knew Ai was a threat to their business model since they appeared so heavily invested in Ai by acquiring deepmind. turns out maybe they were not.

[@ran160]

I don’t think they knew how popular it would become that’s why they never released any ai chat bots for beta testing.

[@shake6321]

@@ran160 They had to have known at least last year when GPT 3 launched. or even the year before when we had gpt 2. Chat GPT is just more popular because it was hyped more by the powers that be. I think Google has too conservative and really thought they could keep Ai in check.

[@mickeywebb6850]

Oh, yes, Deepmind. I almost forgot about them. Hassabis must be worrying.

[@shake6321]

@@mickeywebb6850 i don’t think hassabis cares. that’s the issue. he is just an Ai nerd. i assumed Google was using Hassabis to actually build products they could ship. i guess i was wrong. Google had a massive head start and was outflanked by a competitor based purely on politics and risk aversion. the same will happen to Amazon as well. an Ai that knows the entire internet and can shop for you will be just as destructive to Amazon as chatgpt is to Google.

[@HubertHeller]

Great content! But what is that annoying background sound?

[@staninjapan07]

Fascinating and rather Sky-net scary! You do such a great job of making the videos both captivating and concise. Not a common skill!

[@leobard7767]

Billy's dad was the chief Eugenicist of the world. They hate us. This power in their hands is an awful prospect for the ' useless eaters' [as they call us].

[@JeffNaef]

Very nice work again! I am just always amused when people say things like "ChatGPT can be very human-like, but sometimes it produces illogical or outright false statements". Yeah that just makes it more human-like in my book.

[@jingbot1071]

👉👉

[@Raghu.5301]

Very nice

[@barringtonmorris90]

ChatGPT will further take the thinking out of human beings. Impressive.

[@Tubeytime]

Wetware is so last era

[@MojalefaTsotetsi]

The beginning had me teary though. Just the innocence and the track doesnt make things better. WHERE DID WE GO WRONG?!?

[@kindofanmol]

A prototype version of Bing's new chatGPT interface just leaked and it looks REALLY good. Exactly what we want.

[@EvertvanBrussel]

What? Where? Can you give a link to anything?

[@EudaderurScheiss]

i just defined what google needs to do in a previous comment and y thats pretty close to what i meant. lets see what google comes up with, they have the data and resources for coming up with a better working solution.

[@supermaniac5]

I honestly don’t mind sifting through information at times to find exactly what I’m looking for. I feel that I can learn more that way. Reading experiences and sometimes knowing what doesn’t work is good information too. I’m hesitate to jump aboard a future with too much AI. Most things in life with artificial in the name are not good for you.

[@user-fk8zw5js2p]

True, true. Too much of anything is not good for you. But, I foresee that eventually we will all have personal AI assistants or be left behind. So, maybe you could have yours find a few results for you to review instead of just finding the "best" result.

[@paul1979uk2000]

Sometimes it's fine but in most cases, we just want a direct response to whatever we want to know, this is where ChatGPT could be a game changer and it goes far beyond that in being able to assist you in any given work you are working on. It's easy to understand why Google would fear this, because why would we use Google if an alternative service gives us the information we want and a lot quicker and can even assist on any given project to speed up the process. Traditional search engines are good for what they are but we still have to do all the donkey work to get things done or to get the information we want, A.I. can turn that on it's head and make us all a lot more productive in life.

[@filiecs3]

It looks like that Microsoft is intending that search queries will cite their sources, so you should be able to visit the links provided if you want more specificity.

[@tommyt7555]

Perhaps you would learn more by going to the library and checking out books that are loosely related to what you’re looking for. You’ll read a lot of Supplementary material. You’ll learn a lot about a plethora of topics. But are you gonna do it?

[@samduss4193]

To be honest i tried GPT and i think it is nice but I feel more confortable using just google cause you target you can get the source and also if u dig you can also get some details that are nice to know.. instead of just dealing with a GPT

[@abangwu]

Dagogo really cataloging & presenting humanity hurtling towards its inevitable end with sweet tunes as backing track. I dig it.