The fact that he is interviewing a senior editor of Forbes two weeks after doing a piece about the Forbes 30 under 30 is what we in the business call „based af“
I mean it's kind of dumb too, that's like asking whether new RTX feature is overhyped and then asking a bartender, an accountant and a basketball referee, but not anyone who does anything related to games or graphics. I think it would have been better to ask AI researchers or people who cover AI professionally.
@@TheManinBlack9054the people who cover ai professionally are like , the absolute worst people to ask. They are literally the ones doing the hyping. That is like asking Mac Rumors if the iPhone is over hyped.
@@hastyscorpion so let's ask an accountant who have no idea the new iPhone have even come out? And with all due respect you are comparing science journals with a low-class, specifically hype-dedicated rumors magazine. In that case it would have been closer to compare it to something like MKBHD or Wired or The Verge. Now that would have been a fairer comparison, but what you have there is a very disingenuous comaprison. I think its important to listen to the experts first.
@@TheManinBlack9054 I agree, we should definitely take this topic very seriously and scientifically, there’s no room to ask for public opinion of the uneducated masses /s
Seeing venture capitalists throw millions upon millions of dollars at startups who are grossly overvalued simply because they make AI, reminds me a lot about Dotcom.
According to the GoodWork dictionary of business terms, the word “help” is defined as: “Something beta business weaklings call for when unable or incapable of completing work themselves.” I refuse to accept any other definitions. Thank you.
It is kind of amazing how utterly non-comital and inherently speculative your economist expert's analysis was. He was the perfect personification of how hype culture is allowed to perpetuate, particularly by experts themselves. You asked him if a thing was, and his response..."maybe." Absolutely hilarious!
Sometimes its better to err on the side of caution then boldly proclaim something that you are not sure about. Life is full of nuance and things we do not yet know.
Well, you see, these guys live and die by prestige. If they make an unambiguous affirmation, they could be proven wrong by the tides of time, and that one blunder could prove a tragedy for their reputation and career. It's better to answer "maybe" to any and all questions, and avoid being proven wrong and thus lose prestige.
HuggingFace , maybe not actually worth the valuation, but is actually a good company that has been democraticizing language models since way before ChatGPT
“Not everyone should be developing the same thing side by side” If we were standing in 1995, then absolutely every company had to make a website over the next 15 years. That being said, not all companies need to reinvent HTML. Companies should use AI not reinvent it, and I think 95% of companies will see direct benefit from integrating AI over the next decade.
With the current chat AI, simply merging your knowledge base / customer service Q&A with AI will be beneficial. They are just Google search with normal speech input and output. Art AI are only good with brainstorming right now unless you can make them to work on your uploads, like adding Dan to your wedding photos. Voice AI is a huge threat as it enhance Deep Fake result and our legal system does not have the means to verify the video. General AI is not coming anytime soon.
We don't need HTML, we already have gopher. Wait, that doesn't work. Gopher is the protocol, so: We don't need HTTP. It can't do half the things Xanadu can do. And we don't need HTML either. ROFF can do much more, with a much simpler parser. Eh, who cares about Neuromancer-style 3D visualisations of hypertext relationships in gopherspace, who cares about backlinks, versioning, or clients which are also authoring tools, when instead you can have thin clients running in a virtual machine defined by a monopoly. The thing about AI is that as long as there is no AGI, you will need to keep reinventing AI, in the same way that you need to write new software for new applications. HTML was a single application: Writing and rendering hypertext. AI can be applied to lots of different things, and each type of application requires a different AI architecture. With AGI you still need to train custom models for your applications, and that is still tricky because different applications may still require different types of AGI. But at least you don't have to design them from the ground up. It's more like using an existing scripting language than designing a custom domain specific language. The actual application of AI, as it is, is in using the models. And that is not as straightforward as using a command in the shell. ChatGPT has absorbed a lot of knowledge, to the point that it can answer questions, sometimes even correctly, but it is not a search engine. For image synthesis and editing you need a completely different type of AI again. And that's just the most hyped models right now. There is also speech recognition, which is usable, but still far from solved. There is face recognition, which works fairly well. There is speech synthesis which used to be solved until people began using AI for it. There is also style transfer in pictures and in music. But what doesn't get talked about in all this hype is how AI has made gene sequencing possible. The human genome project would have been impossible without AI, and now it is being applied to the whole biosphere. It is also applied to the stock market, but that doesn't get talked about much either. There will always be more that can be done with AI. But the sobering truth is that AI has been with us and in practical use for decades. The current hype is about trying to find something, anything, that can be monopolised. But computers are universal machines: What one can do, anyone can do. Just like how every computer can serve web pages, share files, compile code, and deliver e-mail. Or mine bitcoins in the webasm sourced from GitHub by an indirect dependency of the bootstrap code of a thin client running in your Chrome-based browser.
AI still no open for everyone, they will reinvent it until is democratic enough. People don't want corporations to hold it but others think id dangerous to allow everyyto have it, the truth is we don't know
@@ishathakor Its pretty helpful for at least some task involved in almost every job. I bet that over the next 15 years then all computer based jobs will be more productive because of ai-powered software, as a direct result of the software being ai-powered.
Would've liked to see you interview a computer scientist when talking to the expert. But I'm glad you didn't interview an "AI expert" (whatever that means) considering they'd probably have a billion dollars riding on whatever they say. As a nerd (and comp sci grad) myself, I think Stephen Wolfram's take on AI is the best. Revolutionary? Maybe. A new and helpful tool like Google was? Yes. And yea, I think AI is overhyped even before software developers. It's very similar to crypto. Eventually, it *could* be pretty revolutionary but for now it's just ripe for making some fat stacks.
So far many of the AI experts that i have seen were very much honest and direct, such as people from MLST or Rob Miles or Bengio and so on and so on. if you dont trust them ask someone from academia. Listen to the experts first. I think the importance of AI and its revoltuonary potential that is already starting to show is much more than crypto's, so its more like underhyped really.
@@TheManinBlack9054because it extorts value from artists without their permission or compensating them. There is also the fact that having a medium mainly done for the sake of human expression being replaced by mass producing machines, is also pretty dystopian.
I work at an autobody shop and we're starting to see preliminary estimates being written by insurance AI...The estimates they shit out are absolutely terrible, making mistakes not even a complete noob would make
If you even need to ask, the answer is always YES Also quite sad that they'd rather buy into hype than actually SOLVE PROBLEMS if they're so flush with that greenback!
i dont comment much on here but your videos are SO UNDERRATED!! great production, i love the diverse opinions presented and your jokes make me giggle!! just wanted to comment to thank you
I read the book "A random walk down wall street". The book mentions the different types of booms that happen in history. Electronics Boom 1960s --> 1970s The Nifty Fifty a.k.a tech companies --> 1980s Biotech boom --> Dot Com Boom 1990s --> 2010 crypto boom --> 2020s AI. What boom/hype do you think will happen next?
Please do a video about why the financial year had to be “different” by starting/ending in the middle of the actual year. Actually a video about why accountants are generally out of touch with being a human would be nice too
would have liked to see interviews with experts on the computer science side of things instead of journalists and economists who don't ultimately understand the technology or ethics. usually love these kinds of videos but this one falls really flat for me.
Also remember that 99.9% of people who have an opinion on this don't understand what they are talking about. Even the people who built the AI don't entirely understand it yet.
The only problem I have with this video is the journalist they decided to interview for this. Forbes has a pretty terrible track record when it comes to business topics.
> Video about a new technology > No insight from specialist in said technology They didn't even try, well at least the economic side of things was dealt with
It is definitely over-hyped by the corporations. They spend hundreds of billions of $$ and agree to regulations because they just want to party a little! 👯
I doubt any of these AI companies are releasing their best work. They're probably only releasing the stuff that's "good enough" to get investors excited, but not too good as to alarm regulators. They would likely be improving the best version of their products in private without being weighed down by far-reaching regulation, waiting for the day when the hype has died down and AI work becomes normalised. Even if it becomes regulated, I doubt they would comply. Facebook has proven by example that a company can break the rules many times over, paying relatively small fines, and keeping most of their ill-gotten gains.
I am currently a grad student pursuing a Master's degree in Artificial Intelligence at one of the top engineering schools in my state. My educated opinion on this topic is as follows: Yes, but actually no.
Im a theoretical physicist and have worked on several quantum computer AI projects. The answer is yes, AI is over hyped, unfortunately it can only ever output things we already know, it wont come up with a cure for cancer for example. It might be good for summarizing large blocks of text however.
imo. It's kind of curious how even this channel would ask the AI question but not really talk to actual AI experts. I think the forbes editor, the economist and the educator provide useful opinions. But I wish folks like Timnit Gebru would get more exposure on these matters. The thing with AI is, it's not true that general AI is true, but we do have a thing that can calculate a bunch of bits out of existing bits and make the bunch of bits look very convincing as if they were new bits but not really.
Yeah as a software engineer this also confused me. My biggest problem with "AI" these days is how often its capabilities are wildly overstated by marketing departments and reporters who don't technically know wtf is even going on. All these tech terms constantly get thrown into a blender and come out as wild speculation that often kind of ignores that there isn't any guarantee that these things will keep linearly getting better forever.
True, so far it is not AGI, but that is quite a big goal and when it happens we might not even be alived to know it, in a very bad way. So its good that its not AGI yet considering we havent really solved the alignment problem. But thats exactly the problem AI is not as much as it is overhyped, its underhyped since it has a very big potential to progress rapidly which it already does and we might not be ready for what comes next.
I just wanna say there was a AI machine built in 1950's that could process things 4500 times faster than the human brain ever could. It was a called a calculator.
@@TheThreatenedSwan Well, the thing is, ever since we've built a machine that could compute data, we've not really been sure where to draw the line between "machine" and "machine that thinks".
So we just need a series of comprehensive large-scale studies to provide accurate data to determine what the proper level of hype is and get everyone to agree to implement a plan to achieve the correct the hype level?
Wait did he just go the whole video speculating about the value of generative ai and then not interview a single computer scientist or statistician who might actually have a chance of understanding the technology? Oh you mean that's why the question exists about whether or not it's over hyped, because nobody actually cares about the science anymore since it's too hard? Oh, ok well that sucks.
It makes you wonder whether this'll help or worsen mental health problems as AI travels towards the entertainment and the loneliness sectors. Or maybe it'll arrive, be perceived as weird by the normies for a while, until finally it becomes the new normal? In either event, this'll do Hell for the dating pool (even further)
I like to remember Dot Com Bubble. Even with a technology as inarguably significant as the internet, it's still possible to get too hyped too early and blow a ton of money.
Hi Good work, you mentioned you are from Mass. I was wondering, are you from the south shore? You strike me as a south shore fellow. Anyway, love the vids, very funny, keep it up.
Remember when businesses were asked how they were going to implement crypto into their business plans? When was the last time you heard the term bitcoin, and it not be a scam
I would unironically buy a Good Work dictionary of business terms
edit - this vid has made me insecure about being a tatless weenie
That edit needs some elaborating
Some of these definitions are funny.
@@TCt83067695 It's a reference to the text on the screen at 4:16 min
I’m loving Toomey’s dedication to finding the ugliest NYC street backgrounds to stand in front of
we don't have to go far
@@GoodWorkMB😂😂
@@GoodWorkMB not with that attitude
Hope he keeps an eye out for CHUDs.
Strongly concur. Please. Do. Always.
The fact that he is interviewing a senior editor of Forbes two weeks after doing a piece about the Forbes 30 under 30 is what we in the business call „based af“
we thank Alex for crossing enemy lines to do this covert discussion with us
I mean it's kind of dumb too, that's like asking whether new RTX feature is overhyped and then asking a bartender, an accountant and a basketball referee, but not anyone who does anything related to games or graphics.
I think it would have been better to ask AI researchers or people who cover AI professionally.
@@TheManinBlack9054the people who cover ai professionally are like , the absolute worst people to ask. They are literally the ones doing the hyping. That is like asking Mac Rumors if the iPhone is over hyped.
@@hastyscorpion so let's ask an accountant who have no idea the new iPhone have even come out? And with all due respect you are comparing science journals with a low-class, specifically hype-dedicated rumors magazine. In that case it would have been closer to compare it to something like MKBHD or Wired or The Verge. Now that would have been a fairer comparison, but what you have there is a very disingenuous comaprison. I think its important to listen to the experts first.
@@TheManinBlack9054 I agree, we should definitely take this topic very seriously and scientifically, there’s no room to ask for public opinion of the uneducated masses /s
Dan Toomey might just be the best investigative journalist on the block right now
The most entertaining one at the very least
He IS the block right now! 👍
on that block of NYC
and those blocks include some of the worst in Manhattan
certainly in the neighborhood
This channel, along with Dan Toomey, has very quickly become my absolute favorite. The analysis, the humor, and the delivery is just perfect.
Same
Seeing venture capitalists throw millions upon millions of dollars at startups who are grossly overvalued simply because they make AI, reminds me a lot about Dotcom.
True, but there might be a few more amazons and paypals in there. Just a bet.
VCs are dumb silver spoon bros. They don't understand anything about what they're betting on.
@@TheManinBlack9054fucking hopefully not
We’re still in’t…. Always have….
And Crypto... And Metaverse... And...
According to the GoodWork dictionary of business terms, the word “help” is defined as: “Something beta business weaklings call for when unable or incapable of completing work themselves.” I refuse to accept any other definitions. Thank you.
Overhyped? Allen Iverson was one of the greatest basketball players ever.
we talkin bout practice?
Ain't even taking about the game
You won😂
Fact
this wins
The connect 4 was the first thing I noticed in that shot, the joke absolutely folded me over 😂
"AI can be both overhyped and properly hyped at the same time." A really high density statement on the topic that I've heard in a while.
in biotech it's kind of becoming a solution for all problem, whilst we still struggle to get enough data to actually implement these technologies
“I *hate* nuance” well said
Spoken like a true journalist
huge fan of what my hair was doing and literally no one told me
We just see the sick tats on your forearms.
It is kind of amazing how utterly non-comital and inherently speculative your economist expert's analysis was. He was the perfect personification of how hype culture is allowed to perpetuate, particularly by experts themselves. You asked him if a thing was, and his response..."maybe." Absolutely hilarious!
Sometimes its better to err on the side of caution then boldly proclaim something that you are not sure about. Life is full of nuance and things we do not yet know.
so true. the forbes guy had a way more thoughtful take and put his argument better. cracks me up
Well, you see, these guys live and die by prestige. If they make an unambiguous affirmation, they could be proven wrong by the tides of time, and that one blunder could prove a tragedy for their reputation and career. It's better to answer "maybe" to any and all questions, and avoid being proven wrong and thus lose prestige.
@@alexanderthedude5474”yes and or no but. maybe” is equally noncommittal, just by committing to everything at the same time
HuggingFace , maybe not actually worth the valuation, but is actually a good company that has been democraticizing language models since way before ChatGPT
not saying it's not! but certainly there's hype to be found
Can’t believe this channel isn’t bigger. God tier content
These videos are crack for my ADHD brain
Agreed, and they pair well with the actual crack I also consume
I like the detail of blurring out the brand of the cup.
Good Work! 0:4:16
i love the dedication put into these videos and the fact that this channel exists in the first place :)
I’m consistently surprised by how underrated this channel is. I wake up everyday hoping that Dan posts a new video
I can't believe how consistently witty and clever these videos are
Dan is the 🐐
Also I hope you find a playdate
The blurred Dunkin Donuts drink in the back is killing me. 😂
Came for the jokes, stayed for the economic analysis. Good work.
“Not everyone should be developing the same thing side by side”
If we were standing in 1995, then absolutely every company had to make a website over the next 15 years.
That being said, not all companies need to reinvent HTML.
Companies should use AI not reinvent it, and I think 95% of companies will see direct benefit from integrating AI over the next decade.
With the current chat AI, simply merging your knowledge base / customer service Q&A with AI will be beneficial. They are just Google search with normal speech input and output. Art AI are only good with brainstorming right now unless you can make them to work on your uploads, like adding Dan to your wedding photos. Voice AI is a huge threat as it enhance Deep Fake result and our legal system does not have the means to verify the video. General AI is not coming anytime soon.
We don't need HTML, we already have gopher.
Wait, that doesn't work. Gopher is the protocol, so: We don't need HTTP. It can't do half the things Xanadu can do.
And we don't need HTML either. ROFF can do much more, with a much simpler parser.
Eh, who cares about Neuromancer-style 3D visualisations of hypertext relationships in gopherspace, who cares about backlinks, versioning, or clients which are also authoring tools, when instead you can have thin clients running in a virtual machine defined by a monopoly.
The thing about AI is that as long as there is no AGI, you will need to keep reinventing AI, in the same way that you need to write new software for new applications. HTML was a single application: Writing and rendering hypertext. AI can be applied to lots of different things, and each type of application requires a different AI architecture.
With AGI you still need to train custom models for your applications, and that is still tricky because different applications may still require different types of AGI. But at least you don't have to design them from the ground up. It's more like using an existing scripting language than designing a custom domain specific language.
The actual application of AI, as it is, is in using the models. And that is not as straightforward as using a command in the shell.
ChatGPT has absorbed a lot of knowledge, to the point that it can answer questions, sometimes even correctly, but it is not a search engine. For image synthesis and editing you need a completely different type of AI again. And that's just the most hyped models right now. There is also speech recognition, which is usable, but still far from solved. There is face recognition, which works fairly well. There is speech synthesis which used to be solved until people began using AI for it. There is also style transfer in pictures and in music.
But what doesn't get talked about in all this hype is how AI has made gene sequencing possible. The human genome project would have been impossible without AI, and now it is being applied to the whole biosphere. It is also applied to the stock market, but that doesn't get talked about much either.
There will always be more that can be done with AI. But the sobering truth is that AI has been with us and in practical use for decades. The current hype is about trying to find something, anything, that can be monopolised.
But computers are universal machines: What one can do, anyone can do. Just like how every computer can serve web pages, share files, compile code, and deliver e-mail. Or mine bitcoins in the webasm sourced from GitHub by an indirect dependency of the bootstrap code of a thin client running in your Chrome-based browser.
AI still no open for everyone, they will reinvent it until is democratic enough.
People don't want corporations to hold it but others think id dangerous to allow everyyto have it, the truth is we don't know
but like.... not everyone needs ai??
@@ishathakor Its pretty helpful for at least some task involved in almost every job. I bet that over the next 15 years then all computer based jobs will be more productive because of ai-powered software, as a direct result of the software being ai-powered.
Would've liked to see you interview a computer scientist when talking to the expert. But I'm glad you didn't interview an "AI expert" (whatever that means) considering they'd probably have a billion dollars riding on whatever they say. As a nerd (and comp sci grad) myself, I think Stephen Wolfram's take on AI is the best. Revolutionary? Maybe. A new and helpful tool like Google was? Yes.
And yea, I think AI is overhyped even before software developers. It's very similar to crypto. Eventually, it *could* be pretty revolutionary but for now it's just ripe for making some fat stacks.
So far many of the AI experts that i have seen were very much honest and direct, such as people from MLST or Rob Miles or Bengio and so on and so on. if you dont trust them ask someone from academia. Listen to the experts first. I think the importance of AI and its revoltuonary potential that is already starting to show is much more than crypto's, so its more like underhyped really.
An AI expert would be a PhD in machine learning working at Microsoft. It's not Marc Andressen, he's an expert on having money.
@@TheManinBlack9054 cryptocurrency obviously had zero importance from the beginning. It's just fake money.
@@weird-guy why is AI art horrible and how is it uncanny?
@@TheManinBlack9054because it extorts value from artists without their permission or compensating them. There is also the fact that having a medium mainly done for the sake of human expression being replaced by mass producing machines, is also pretty dystopian.
I work at an autobody shop and we're starting to see preliminary estimates being written by insurance AI...The estimates they shit out are absolutely terrible, making mistakes not even a complete noob would make
THIS CHANNEL FEEL'S NOSTALGIC, THE WAY IT USED TO BE THIS GUY DOES IT ALL THANKS TOOMEY
If you even need to ask, the answer is always YES
Also quite sad that they'd rather buy into hype than actually SOLVE PROBLEMS if they're so flush with that greenback!
I mean, it's usually the opposite way, lol. If there's a question in the title of the video then the answer is no.
@@TheManinBlack9054Glad you're here to teach us about nuance in a world built to destroy rhetoric 😊
> do you have money?
YES
> invest in AI
why?
> YES.
:D
@@effexon Reminds me of finance bros back in 2020-21 trying to convince all their friends to buy bitcoin
@@Patrickstarrrrr69they still are.... gold and silver are other evergreens :P
3:05 "Really likes Connect4" HAHAHA I AM WHEEZINGGGG
U had Kyla come on the show?!
WOW
DAYM
Favorite part: Jason Abaluck really likes Connect 4. Dan Tooley trying not to mention the Connect 4 thing
😅
really likes connect four
i dont comment much on here but your videos are SO UNDERRATED!! great production, i love the diverse opinions presented and your jokes make me giggle!! just wanted to comment to thank you
As a software engineer who did his thesis on AI, yes it is so overhyped. Journos can't be more dramatic even if they tried!
The first thing I learnt from one of my teachers was that computers will only ever do what you tell them to do no matter how 'smart' they become.
Journalists are just business funded by venture capitals, that's why.
@@MasalaMan yes, that's basically it
As a person who works in AI (AI engineer) I agree with you haha
@@MasalaMan well, you learned wrong i guess. AI is not just your average calculator program
I read the book "A random walk down wall street". The book mentions the different types of booms that happen in history. Electronics Boom 1960s --> 1970s The Nifty Fifty a.k.a tech companies --> 1980s Biotech boom --> Dot Com Boom 1990s --> 2010 crypto boom --> 2020s AI. What boom/hype do you think will happen next?
Please do a video about why the financial year had to be “different” by starting/ending in the middle of the actual year. Actually a video about why accountants are generally out of touch with being a human would be nice too
I miss the yellow descriptions of the guest and Dan. Sure, there were two instances but would love more! Great video though 👌
would have liked to see interviews with experts on the computer science side of things instead of journalists and economists who don't ultimately understand the technology or ethics. usually love these kinds of videos but this one falls really flat for me.
Alex Konrad the Senior Editor at Forbes managazine carrying this show again.....love you
Next video idea: Are The Hollywood Executives Overpaid?
as a data scientist, i thoroughly enjoyed this and am definitely going for a career change soon 😂
change to what? And why change?
@@iceman7179 It might be a joke.
Also remember that 99.9% of people who have an opinion on this don't understand what they are talking about.
Even the people who built the AI don't entirely understand it yet.
Should talk to an AI expert. Likely true that none of these people know anything meaningful about AI beyond the hype.
it’s an AI experts job to hype up AI, tho. if AI is overhyped then their career is at risk
hugging face is not a random AI startup
The only problem I have with this video is the journalist they decided to interview for this. Forbes has a pretty terrible track record when it comes to business topics.
i think it's instructive to see charlatans in action
> Video about a new technology
> No insight from specialist in said technology
They didn't even try, well at least the economic side of things was dealt with
You almost got the joke lol
@@snarf1504 what joke? The joke that he is misnforming people by asking the people who know nothing about the topic?
@@TheManinBlack9054 Exactly, this is a comedy channel 🤣
I've seen the last few videos you've made, you're going to hit million subs. You're improving at such a rapid pace and it's awesome to see
Good video, and always love seeing Kyla Scanlon pop up!
A masterpiece of investigative journalism
Keep doing the good work!
0:47 great dictionary
the hype around AI just reminds me of when everyone kept talking about the metaverse like a year ago
This is such a good point!
Lol exactly.
"It's going to change the world forever, bro!"
Except adoption of AI has the numbers the metaverse never had. It's popular, it actually does stuff people want, and it has a low barrier to entry.
Everybody got a piece to say but no one watches the video or wants to hear what is being said
Everyone said the metaverse was a garbage idea right from the start tho. Basically the opposite of ai
It is definitely over-hyped by the corporations.
They spend hundreds of billions of $$ and agree to regulations because they just want to party a little! 👯
How do you find all these beautiful locations?
this channel needs more attention
Yes and. Or no but. Either one.
I doubt any of these AI companies are releasing their best work. They're probably only releasing the stuff that's "good enough" to get investors excited, but not too good as to alarm regulators. They would likely be improving the best version of their products in private without being weighed down by far-reaching regulation, waiting for the day when the hype has died down and AI work becomes normalised. Even if it becomes regulated, I doubt they would comply. Facebook has proven by example that a company can break the rules many times over, paying relatively small fines, and keeping most of their ill-gotten gains.
No ai guy in this video 😢
A.I. deez nutz lmaoo
ok that was good
Gottem
Instructions unclear.
This channel is so underrated
0:00-0:06
Just a banger intro, immediately hooked.
here for Kyla!
I love Kyla Scanlon! She also has great tattoos.
I am here because Bricky mentioned you.
and god am I glad I am here.
The only news channel i wanna watch
I need to know why dua lipa loves goodwork so much
Love Kyla's work!
Giving you a like before I've even watched the video. Good work!
thank you for supporting journalism
I am currently a grad student pursuing a Master's degree in Artificial Intelligence at one of the top engineering schools in my state. My educated opinion on this topic is as follows:
Yes, but actually no.
Im a theoretical physicist and have worked on several quantum computer AI projects. The answer is yes, AI is over hyped, unfortunately it can only ever output things we already know, it wont come up with a cure for cancer for example. It might be good for summarizing large blocks of text however.
How exactly is it overhyped @@user-ez7ls2du9c
6:16 this man's work ethic is on POINT
/respect
To this day, Toomey is still trying to not bring up that Connect 4 thing
imo. It's kind of curious how even this channel would ask the AI question but not really talk to actual AI experts. I think the forbes editor, the economist and the educator provide useful opinions. But I wish folks like Timnit Gebru would get more exposure on these matters. The thing with AI is, it's not true that general AI is true, but we do have a thing that can calculate a bunch of bits out of existing bits and make the bunch of bits look very convincing as if they were new bits but not really.
Yeah as a software engineer this also confused me. My biggest problem with "AI" these days is how often its capabilities are wildly overstated by marketing departments and reporters who don't technically know wtf is even going on. All these tech terms constantly get thrown into a blender and come out as wild speculation that often kind of ignores that there isn't any guarantee that these things will keep linearly getting better forever.
Stochastic parrots :)
@@jaiveersingh4981 i dont think you understand what you are talking about.
True, so far it is not AGI, but that is quite a big goal and when it happens we might not even be alived to know it, in a very bad way. So its good that its not AGI yet considering we havent really solved the alignment problem. But thats exactly the problem AI is not as much as it is overhyped, its underhyped since it has a very big potential to progress rapidly which it already does and we might not be ready for what comes next.
@@TheManinBlack9054 Actually, I think I do. I work on AI at an extremely famous AI-enabling company with a bright green logo :)
Thank you Dan for reminding me why I never want to move to NYC
I just wanna say there was a AI machine built in 1950's that could process things 4500 times faster than the human brain ever could.
It was a called a calculator.
now its michael jackson selling ya gucci brand vacuum!
Haha exactly
That's not intelligence. AI is also not general intelligence.
@@TheThreatenedSwan Well, the thing is, ever since we've built a machine that could compute data, we've not really been sure where to draw the line between "machine" and "machine that thinks".
But it requires you telling it what to do step by step, it’s not intelligent
You trying not to mention the connect four thing to the Yale professor made me laugh my ass off😂
I couldn’t take in a word that gentleman was saying throughout the connect 4 saga.
So we just need a series of comprehensive large-scale studies to provide accurate data to determine what the proper level of hype is and get everyone to agree to implement a plan to achieve the correct the hype level?
can you please do a video regarding Investment Banking? Thanks. Love your videos
The connect4 bit is soo good😂😂
Wait did he just go the whole video speculating about the value of generative ai and then not interview a single computer scientist or statistician who might actually have a chance of understanding the technology? Oh you mean that's why the question exists about whether or not it's over hyped, because nobody actually cares about the science anymore since it's too hard? Oh, ok well that sucks.
idc about the climate, playing with AI kicks ass. I'm tryna figure out a way to replace people's jobs, wish me luck!!
It makes you wonder whether this'll help or worsen mental health problems as AI travels towards the entertainment and the loneliness sectors. Or maybe it'll arrive, be perceived as weird by the normies for a while, until finally it becomes the new normal?
In either event, this'll do Hell for the dating pool (even further)
I like to remember Dot Com Bubble. Even with a technology as inarguably significant as the internet, it's still possible to get too hyped too early and blow a ton of money.
Probably not overrated
its totally overhyped unless you are a programmer... us programmers are fucked
Strange that the "expert" you spoke to was just... a finance guy. That's not an "expert" on AI lol
Loves connect 4, give your editor a raise
Every specific application is overhyped but it overall is underhyped
This guy show outstanding dedication to every video.
love this channel! wondering if my subscription to morning brew triggered the recommendation algorithm LOL
Great work on the presentation! 😊🙏
Can't wait for the GoodWork_ Dictionary of Business Terms paperback edition.
But did you ask ChatGPT?
Your scenery composition never disappoints LMAO. Love your page.
4:47 she is so gorgeous.
@goodworkMB - But did you buy the vacuum or not?
Hi Good work, you mentioned you are from Mass. I was wondering, are you from the south shore? You strike me as a south shore fellow. Anyway, love the vids, very funny, keep it up.
Hold up say Nvidia again
Nah-Vidia. We pronounce everything in MLA Format, although I understand many are used to the Chicago style
Remember when businesses were asked how they were going to implement crypto into their business plans?
When was the last time you heard the term bitcoin, and it not be a scam
I am so glad to have chanced upon this channel. It’s hilarious 😅😅😅
th blooper reel was brilliant. Great stuff!