It will be quite easy to replace GPs in UK. Whenever you go to GP with any problem, first they tell you to stay hydrated and if that doesn't work, they tell you get paracetamol. ChatGPT can easily replace 80% of GPs.
Not only that, but GPs routinely make treatment recommendations that are not evidence-based and even that disagree with clinical guidelines. And they often make recommendations that are biased on gender, race, economic status or physical appearance. But while it may be theoretically easy to replace a doctor, the political reality is a different story. Physicians have protected their monopoly for a very long time, and have successfully insulated themselves from many challenges to the status quo. They won't go down without a very hard fight.
@Moonlight Gamers Just an observation. I don't necessarily judge physicians or even their boards for this. Medical schooling is an issue, as is the way we fund/reimburse physicians for their services. Issues with bias and substandard decision-making occur in many industries and public services. And anyone could be forgiven for wanting to protect their livelihood, especially after spending so much time and money to attain it. Perhaps I'm even overestimating their political power, because the winds of change are already blowing. I don't know about the UK, but in North America, Nurse Practitioners and Physician Assistants have already, for some time, been replacing MDs in many situations. It'll certainly be interesting to see what the fields of medicine, law, design, engineering, architecture, and so many others look like in...5... 10... 15 years. It won't be a question of whether they change or not, it'll just be a question of how dramatically they change.
@Moonlight Gamers No... OP and I were talking about the UK. And when my commentary about NPs and PAs departed the UK, I explicitly said so. I've worked in global public health research for more than a decade--my perspectives on this aren't coming out of left field. I'm getting the feeling that you have opinions about this topic. Rather than just reacting to what I have to say, I'd love to hear your thoughts on the matter.
@Moonlight Gamers And anyone can go on RU-vid and randomly vilify ("such hate") or misrepresent ("you generalized...") or attack ("anyone can claim") other people's perspectives and claims without having any perspectives or claims of their own. It doesn't matter to me whether you believe me about my work experience, and if you're interested in learning more about the topics I discussed, you can easily find more information online, or just ask. If you'd like to discuss the topic further in a meaningful way, I'd be happy to. But if the purpose of your participation in this thread is just to shoot other people down, take it somewhere else, because that's about as boring as boring can be. As for the notion that science evolves--yes, I certainly agree. The field of medicine, like the fields of law, design, engineering, and many others, is poised to undergo significant change. As for whether your opinion counts, I think it should, absolutely. Our healthcare system affects all of us. It's our money paying for it and whether it works well or not is literally a matter of life and death.
ChatGPT Hype vs Reality - is this AI going to take our jobs away? What's the truth about AI? Big thanks to Brilliant for sponsoring this video! Get started with a 20% discount using this link: brilliant.org/davidbombal How do you capitalize on this trend and learn AI? Dr Mike Pound of Computerphile fame tells us how to ride this wave. If you are into cybersecurity or any other tech field, you probably want to learn about AI and ML. They can really help your resume and help you increase the $$$ you earn. AI just become Sentient? And will it take your job? Or is AI just a fantastic opportunity for you to get a better job? In this interview with Dr Michael Pound we discuss hype vs reality and get a quick start guide on how to learn AI. // MENU // 00:00 - Coming up 00:40 - Sponsored segment 02:28 - A.I. Hype // Should we be worried? 03:37 - Amazing but flawed 08:07 - Is it worth it getting into CompSci? 10:02 - Knowing A.I. makes you valuable // Learn A.I. 13:43 - Resources for learning A.I. 15:58 - Should you get into CompSci? 17:35 - Enhancing your career with A.I. 20:16 - The limits of A.I. 24:57 - A.I in academics // How A.I. affects academic work 31:02 - Conclusion // Previous Videos // I challenged ChatGPT to code and hack: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Fw5ybNwwSbg.html The truth about AI and why you should learn it - Computerphile explains: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-PH9RQ6Yx75c.html // BOOK // Deep learning by Ian Goodfellow, Yoshua Bengio and Aaron Courville: amzn.to/3vmu4LP // Courses and tutorials // AI For Everyone by Andrew Ng: www.coursera.org/learn/ai-for-everyone PyTorch Tutorial From Research to Production: www.infoq.com/presentations/pytorch-torchscript-botorch/ Scikit-learn Machine Learning in Python: scikit-learn.org/stable/ // PyTorch // Github: github.com/pytorch Website: pytorch.org/ Documentation: ai.facebook.com/tools/pytorch/ // Mike SOCIAL // Twitter: twitter.com/_mikepound RU-vid: ru-vid.com Website: www.nottingham.ac.uk/research/beacons-of-excellence/future-food/meet-the-team/michael-pound/index.aspx // David SOCIAL // Discord: discord.com/invite/usKSyzb Twitter: twitter.com/davidbombal Instagram: instagram.com/davidbombal LinkedIn: www.linkedin.com/in/davidbombal Facebook: facebook.com/davidbombal.co TikTok: tiktok.com/@davidbombal RU-vid: ru-vid.com // MY STUFF // www.amazon.com/shop/davidbombal // SPONSORS // Interested in sponsoring my videos? Reach out to my team here: sponsors@davidbombal.com Please note that links listed may be affiliate links and provide me with a small percentage/kickback should you use them to purchase any of the items listed or recommended. Thank you for supporting me and this channel!
David! Please consider making some sort of Forum or FAQ space (even building a team of your own for it) where people can reach out to you (or your team) for technical advice and to clear doubts about the devices you've covered in the past and Cybersecurity processes.
thank you so much sir david, i didnt go to university and went work straight after college. learn everything watching your youtube!!! so with all those books, coursera, brilliant you mentioned which order should we go for? or doesnt it matter. my math is only at very average level
Chat GPT can be your virtual teacher. I asked it to make things easier to understand and give me examples etc and then I also confirm things and for that purpose its amazing and save you the pain of googling.
Just be cautious and check the answer yourself. I tried testing how good it was at giving examples of programs to simulate basic circuits/physics to see if it could be trusted, and I found multiple errors in the examples it gave such as not using the proper equations or constants. Maybe I was being a bit too hopeful and asked it something too complicated, but keep in mind its examples may not be *good* examples.
@@Lwyte17 it’s a language model, it’s not been designed or trained for maths and equations and this AI model is sub optimal for those operations anyway. Try something like wolfram alpha that has been specifically designed and trained for those kinds of operations. Also how you build your seed and task prompts has a huge impact on the quality of output.
The manner in which you performed the ad for Brilliant was outstanding. Best I've ever seen/heard. It informed, had "Brilliant Sponsored Segment" up for over a minute yet didn't distract the viewer. Really good job! I almost forgot that segment was an ad :-)
a virtual teacher who likes to lie sometimes. Ask him to list you historical figures or grill him about some theory and sooner than later it will start to make things up.
Except you can't trust anything it says, and you weaken yourself the more you use it. The best teacher teaches you how to think; some trash ml can NEVER do that.
Really great to hear the other side of the coin. Most people are freaking out about this, and this gentleman was pretty chill and not worried at all. I think he's correct, for the most part, but the 'next iteration' of the software, although mentioned, wasn't emphasized enough. I think maybe two more iterations of this tech we'll be talking about an almost different beast. And that's in five years or less. Zip, zap, zop, we could have our first AGI in no time.
i use chatgpt for explaining commands and programming functions and researching things in minutes that would take hours using google. it is a tool. i cannot wait to see what ai will be in 5 years
chatGPT is the greatest general computer software assistant tool than I have seen in my 40 year career. It won't take my job BUT it can make me a hell of a lot more efficient in it.
RU-vid would not be what it is without you David. I would have been a passionate supporter of your channel if we hadn't run into payment difficulties in Iran. Hope everything is going well for you. ❤
One thing I think is important going forward is not forcing everyone to get programmer jobs to survive, but to allow us to live and work careers that we want without being forced to use the next big thing in tech. This is why I personally dislike when people get angry at people for not wanting to use AI.
@@idkwhy77 Maybe no one in particular is forcing oneself, however the current function of society as a whole incentivizes it. It is imperative that we make a civilization to allow ourselves to live the lives we want, AI or not, lest we slip into a world where despite our great advancements, we are still following the primitive 'adapt or die' moniker.
Exceptional chat. David all your questions were excellent and Dr Mike Pound answers were brilliant as well. I always enjoy your videos and Computerphile too. Thanks!
Very much enjoying this piece. My best use of chatGPT is for analogy. I broke excel. Because of the size of the dataset, 90 minutes to recalculate. Vlookup was the culprit. So asking chatGPT, "What is the pandas equivalent of excel's Vlookup function?" Quick answer from chatGPT gave the structure I needed. Quicker than reading documentation. Analogy is a valuable use case.
I think that the reality of needing knowledge of CS is a requirement more than ever because of this breakthrough. This is entirely overlooked currently. I mean, think about how advanced anyone with knowledge of these matters will be compared to other people whom are not usually aware of these things.
@@howardn501 Well that's true. We have the resources to supply people with basic nutrition to prevent them from stealing food but we don't. Some people steal or cheat just because they want to, those people are just fundamentally losers and we aren't referring to them
@@howardn501 you are comparing learning with stealing? the point is, student learning is important and not just getting a good grade, instead just focusing on student's grade to measure their intelligence and knowledge 🤣
Well everyone is unique they got undiscovered specialization in them ,u can't ask Ronaldo to study, or musk to play football. Or maybe a future philosopher to do above stuffs 😅
Today I found a very interesting problem which is a problem for ChatGPT. This problem is called man. :-) This program creates statements and program code based on what the user knows. If the user doesn't use the right vocabulary, ChatGPT returns not what we expected. It doesn't mean it worked badly, but it will do something we didn't want to have. So there can only be so much that the average educated user will still have to learn something. :-)
Input makes a huge difference. This is something I had to learn, and I used the combination of the errors that were returned in the terminal in addition to the explanation that ChatGPT gives to achieve the desired results.
Well done David and Mike! I appreciate you really solidifying for folks that ChatGPT is going to disrupt everything however like the industrial revolution disrupted work so will this. JOBS are NOT going away their scope will change.
hi I have tested chapgpt with pl/pgsql functions and was suprised how well it coped. chatgpt cannot work with my indexes which means some functions cannot be run because chatgpt would need access to shema indexing. However, I am working on a java bot that allows a connection to schema indexing to see if the functions program better. I think in the future connection bots will be able to scan schema indexs to produce cte and functions that work with postgresql engine via an open connection like an api. The only area I can see chatgpt will never work is the understanding of the right question to ask for pulling data under the select command. This is a human only trait because chatgpt cannot predict what data sets needs to be pulled regardless of schema indexing. I think chatgpt could be good for sql. thank you for the video.
It will and already is replacing jobs. 200 000 big tech jobs gone this year already and we’re only just in February. It’s going to be exponential so watch this space.
Justin James my chatgpt question: what is the future of sql developers and ai taking away jobs chatgpt answer: It is unlikely that AI will completely replace SQL developers in the near future. While AI can automate some aspects of database management and analysis, SQL developers bring valuable expertise in data modeling, database design, data security, and troubleshooting. Additionally, as more organizations adopt AI, there will likely be a growing demand for SQL developers to integrate AI technologies into their databases and ensure that data is properly stored and managed. However, it's important for SQL developers to stay updated on the latest developments in AI and to continuously enhance their skills to remain competitive in the job market. As AI becomes more integrated into various industries, the demand for professionals who understand both AI and SQL may increase.
A one person perspective question on artificial intelligence in the era of “chatgpt”, and the direction of humanity…. I don't want to come off as paranoid here with this comment, however I cannot just sit back without postulating this position/query that I'm not hearing being discussed as we charge forward. forget everything that you imagine about chatgpt and artificial intelligence in general and ask yourself this question (that I'm not hearing bandied about in this new digital era); what level of "back end" monitoring and data collection about YOU is going on? Do we4 really think that this kind of release is benign to the wiles of commerce? Based on your (and the collective) conversations/queries? what digital biography of yourself are you creating (freely) for the overlords (owners) of the artificial intelligence community to use? How will they (or are they planning on) use this digital footprint of YOU, let alone the global concept of what it means to be human? Think for a moment about the footprints we’ve already created through our digital interactions over the past 30 years. We’ve moved from a position of data collection (old style computing) into digital storybooks of us through social media, where we’ve become the product for the modern day, digital marketeers with our ability to blithely (and freely) click on a (yes, I’ve read and accept your “terms of use” policy) without giving it a second thought. the more data about you, that you freely provide, offers up a new era digital playbook, a sort of personal portfolio of who you are, what you do, what you think, where your mind is going..... data science, and the collection of millions (billions) of human thoughts and the unlimited possibilities of both the individual as well as the collective, might just be something that needs a little more thought behind it before we plunge down this rabbit hole of the future (of which we know not a true direction, let alone philosophical outcome).
27:00 - on recognizing common AI phrasing - future possibility: if rapid AI adoption by students as search engines: just as with generational saying (example: cool, yo, Daddio, chill out, smoke show, hot, sick, wicked, brb, lol, um, whoa, etc) human students might pick up these 'familiarity habits'. This could be fun to track over time. I would be interested to know what you consider gpt typical phrasing to be, or how to find out.
Using it as a helper is efficient and allows to get a lot done while also using the knowledge/creativity of the human. I look at GPT as an assistant. I hope in the future there can be other programs made with a similar model as GPT but used for being a Assistant for specific tasks. I guess there is a Google Assistant, but it would be cool to see it fused with GPT. I would imagine people would be extremely efficient because of how many menial jobs they could give over to that assistant.
A programmer take - Not worried that ChatGPT will take my job. People entering the field in a few years would be just fine. My kids generation, I’d be worried; contemplating teaching them programming. I give it ten years. The next big tech are ones that can allow a business owner to make requests like, create for me a webpage using google as a domain on the google platform using stack xxxx with login etc. It’s only a matter of time.
Funny to listen to Mike especially as I have taken a course of his and watch his very good explantions on his RU-vid videos. If you are writing production code you need to write Unit tests and you need all the static analysis tools you can get. ChatGPT can write tests! so always make it write tests...... As for security very few dev can keep up to date, and I think it would waste valuable time, so use Static Analysis tools. There is no such thing as a bad dev, just Devs without the correct tooling.
I have seen that if I ask it to do any coding that is not readily available as a demo on the internet, it just produces incorrect code. The thing with AI is that it is fairly easy to get it to work almost correctly, but nearly impossible to make it work correctly. For example, we have almost self-driving cars which were developed really in a few years time. But after that we never seem to have a car that actually is self-driving.
Really good at coding regular expressions, and other simple well defined context, but I tried to create some python script for Blender, creating some solids, apply materials and render to image, was quite a long iterative work with a lot of errors made, chatGPT get confused on product version, properties of objects and function to use.
Here's one way to think about *using* AI to pass all your courses: If you obtain a college degree by using AI to take all your tests and write all your assignments, then you can be *certain* that everything you are smart enough to do can be done faster and cheaper by a company going to an AI. You, as a person, are bringing no unique benefit to someone who might hire you.
I swear i thought the exact same thing while watching the video, all the degrees that you can easily get using AI and still after that pass the interview of a job you apply to are completely useless, therefore you as an individual are as well (to the company). Your job can easily be replaced by automation and AI.
This is a bit of a circular argument, isn't it :) the benefit you bring in any case is your ability to adapt and solve problems beyond the scope of tests. Knowledge work is about implementing knowledge and not knowing things. It's about practising learning. However, the issue might show limitations of the current education system that relies on answering questions and adapting to the system. Learning to use AI systems for information gathering is certainly useful and might lead to more adaption in certain narrow domains, which provides value to companies.
@@davidw8668 - The observation I'm making seems pretty direct to me. If some student relies on this and only this to get through their classes, then by the time they graduate they won't have any skills that an employer would want to pay for. That's not the same as a student who learns what an AI can (and cannot) do. But they have to do more than just typing in their assignments to the AI and handing those answers in for their homework.
They can easily change the exams to written exams. Just finished a network exam with pen and paper. Chat gpt is a booster for data access if you don't know what to access aka have the concepts in mind and how everything is tied together you won't be able to do very much with it.
Many will regret underestimating chatGPT-3 (not to mention its successors). Best case, it will be the catalyst for higher intelligence, like what Casio did for calculators.
Idk, I had it complete a pretty difficult function using code today. It's not the most advanced thing in the world. But I got it to compare the prices of a company like "Good Chop" and meats by weight per oz, to prices in the grocery store, and got to get totals, price per oz, etc.
hi David, just asked chat gpt how to make coffee with a French press . i know how to make it but tried the example given . coffee wasn't that bad. what i got from the chat was you need to know something about what you are asking and use the information from chat to make it better.
people now care that AI is going to take their jobs? That's hilarious. If anything its well deserved. These very same people were unapologetic when a lot of industry became automated, leaving many unemployed with an obsolete set of skills per their words. Now that artists, musicians, programmers are screaming that AI will become so advance it will script itself without human interaction is simply music to my ears. I find it amazing how people care about losing their jobs, their employment by automated replacement, when it personally affects them.
this whole conversation seemed to me a little bit on the denial side ,specially when he said taht he is not expert in AI and had limited experience? why should i take advice from someone that says that about any subject ? on the other hand i watched an interview with one of the founders of AI and the stuff he is saying they are going to release this year * that they already have) is just incredible , and i have no reason to think he may be lying considering the crazy exponetial advances that I am seeing ...it the tech curve is indeed exponential (wich until this moment seems to be the case ) then yes there is no reason to not actually think that coding wiill be taken over by AI , sooner than later
Just had to try chatGPT. Actually I found a real imaginary complex error 😂 I asked for an example involving complex numbers, and the example given from chatGPT was z:=(30+40*%i) with an explanation that multiplying with %i the imaginary unit vector, would make the angle 40°. Afterwords using angle(z) would return 40 degrees, the imaginary part. But arctan(30/40)~36.87°. Likewise magnitude(z) would return 30, the real part, but from the 3-4-5 right triangle we know it must be 30-40-50, i.e. 50. Making the bot aware of the error resulted in an excuse, followed up by another result, this time correct. Stay alert!
I have run my own virtual tests. You ask it a question about something like accounting questions, and it tends to defer the answer. The software that it generates is very questionable as to wether it will run. Although I have found that it does work well with NumPy if you know what you want the result to be.
Your video was excellent! Take your technology game to the next level with the "Ultimate Gpt Toolbar" for Chrome, complete with voice control for GPT, Word and PDF export options, and over 200 prompts. It had a huge success with 10,000 downloads in just one day.
Everything suggests that the brain works the same way as ChatGPT, the impression that we decide on our thoughts and actions is only an illusion, we just become aware of a functioning.
I think this is the best video in YT which explaining the truth of ChatGPT I test GPT for Mathematical equations for Linear Algebraic related to elliptic curve hehehe, he is still learning but what you give him GPT will give you and I think this is not proper way for AI, and is not strong enough to give me the best solution. I have to tell him what is the correct to be able to be a satisfy answer, BTW ChatGPT has awesome work. It needs time to be perfect and kick out some human jobs, not within this decade, but I think will take it in future!
Some information stated was incorrect, specifically that LLMs with a sufficient number of parameters do in fact create world models within, an emergent behavior
Thanks for this video. I have been thinking about this topic recently. As a 48 year old carpenter looking at a career change into I.T. I've been wondering about the security of programming jobs.
Well look at this way, if AI ends up taking over programming jobs, you’ll always be able to fall back on your carpentry skills, because AI won’t be doing that for a while! What will the guys who’ve only ever done programming do? My advice would be to get into coding and software development for AI. Personally I think AI will replace a lot of jobs, because it’s going to make technology and knowledge workers a lot more productive. There will still be those jobs, but less of them, as each worker will be doing the job of 3 or 5 or 10 workers. However, AI is going to be a huge growth area, and there is a skills shortage in AI right now, so that’s the way to go. Just my opinion, but I think you’re in a better position than a lot of people because you already have skills that AI isn’t going to replace. Good luck.
ChatGPT is interesting. I think it has its strength, and its weaknesses. What those are, that's an explorative process. In my case as a professional Network Engineer, I find it to be pretty spot on when it comes to ascertaining questions I have. Primarily networking related of course, whether that be certain concepts, explaining routing protocols, testing me on subjects, etc. However, where I've noticed it become weak is areas of programming, or scripting. In some instances, it's very useful. I ran into a situation where I needed to mass produce DHCP scopes for a lab environment, and it not only created a viable script in PowerShell, but reverse engineered it so I could learn the concepts of variables, arrays, etc. Something I never thought I could learn. On the other hand, I'll ask it questions, and it'll be so sure of itself even though the script fails. I can poke, prod, and provide error messages, but then it'll recreate it nigh identical only to fail. This is where I've realized sometimes a prompt simply isn't enough. Rather, it performs better with live data, and then can craft a solution around said data. Very similar, yet different. Key detail. All in all though, I find it to be incredibly useful. Programming/Scripting always seemed to be this ethereal concept to me. Similar to high end math. It's why I strayed away from Electrical Engineering because I had such a poor foundation, I didn't think I could cut it. But now, I feel more confident cross-referencing material online, and actually learning programming is now fun.
Sorry for posting an unrelated comment here in this video. I appreciate this video tho and watched it all and gave the thumbs up :) My question is, can you make a video and teach us on how to set up our home network? the whole switch, firewall etc and make it safe? Thank you
Well, my perception of chatGPT is that it's a great model of AI language, but as every technological thing always needs to improve, I realized doing some tests that this AI doesn't give the raw information of something you ask and it doesn't behave like a teacher, explaining in a didactic way - almost drawing for those who doesn't understand -, and about the abstraction of many subjects, it is still far away. It's like copying and pasting the entire summary of a google search. When asking something we have to be careful with the answers that are given, as it is not considered all the pros and cons. If we question a certain subject several times and say that "my understanding of something is correct and the information from the AI is incorrect" the program doesn't know how to work around and substantiate whether something can be correct or not. Just like 7 + 1 which is 8, if we question that the AI information is incorrect then it will be considered that the result 8 is incorrect then.
Definitely a hype. I'm very much convinced that we are going to see a drastic changes in the intellectual property laws in the near future. Current situation is the (hopefully) short lived "wild west" period where AI models are running loose and use whatever data they want to train. I'm convinced this is going to change. I wouldn't be surprised if the use of generative AI would require disclaimer and/or the publication or the used prompt. As for the job security, if your actual job has any meaning, it won't be taken by AI. Were you writing mediocre copy for advertising? I would tell you long time ago to consider career change.
This was really interesting. I am a 50 year old Accountant thinking about a career change to Tech and I am not sure of the pathway to decide upon. This video has given me some thing to chew upon, any advice?
Get into developing AI or applying AI. If you have 30 plus years of accounting and financial management experience there’s no reason at all why you can’t use that and combine it with your new AI knowledge. In fact I would see it as an advantage in terms of AI application. You are probably familiar with financial management software and processes and are aware of the specific pain points and bottlenecks they produce. I am the founder and CEO of an AI startup, looking at some industry disrupting applied AI research. I have a project that we could possibly collaborate on if you are interested, for example using AI to automate invoice data entry and 3 way matching. That’s just one small example.
This is some great stuff! Have you seen the work that David Shapiro has been doing in this same area, especially with his chatbot, Raven? He has been experimenting with some other ideas with compressing context, such as using vectors and graph databases.
CHATGPT can either cut debugging time dramatically or it can mislead you to solutions that look like it works but may not. CHATGPT still requires an experienced pilot for it to be valuable for software development.
So many people want to rush into AI because of recent hype, but huge massive brick is waiting on their way - "hard, boring, useless, frustrating, ...." MATH - that you wanted to avoid in life at all costs. Good luck with that now!
As a tech myself I think AI is great and I don't think it's something that will take people's jobs quite yet. The thing we need to be afraid of is the person that's part of the C level execs that thinks AI is better than a person and gets convinced that it can replace a person. Business minded people don't know how this actually works and think saving a buck is more important. Best of luck to all who have been hit from the recent tech layoffs.
True about the code that ChatGPT produces. I have tested some not so simple code and ChatGPT failed monumentally. It works great on writing essays, letters etc, but not 100% grammatically correct either. Simple scripts, it gets right. But more complexity code, it fails terribly. Anything that has more than a program, that returns a few variables, and calls some functions, and then prints out. Not great results. Sometimes the code doesn't even compile. With debugging some snipbits, I ran into the same issues. Test everything it gives you. I wonder if the people from openai have curtailed it's abilities to deal with fearful reactions from the public.
Hey guys! How big of a game changer is Emad Mostaque's companies going to be? He seems quite confident that his compression software along with A.I. software is going to take over the world. What say yous? Thanks ahead of time.
18 year olds be looking for a career but... I want to switch careers to longevity to become '18 years old' 🤣 There are some amazing startups in this area and AI will boost them I hope..
Just a trivial question as an American...what is the reason why people from England, or surrounding areas always say: "X, Y and Zed"? Instead of just saying "X, Y and Z"? Just curious, if there's a specific reason and where does that come from.
I think Mike is looking at this from an individual point of view. Saying it won't take humans out of the loop.... Well yes it doesn't have to take them all out of the loop to cause a jobs apocalypse. Here's an example. A large org might have a team made up of a business analyst, a lead Dev and a team of 8 junior coders. We are already at the point where those 8 coders become only 2. Then we may reach the situation with a lead Dev but no junior coders needed only AI. You don't need to take all humans out of the loop. Just most of them. And you've saved a vast amount of money. A jobs apocalypse.
