Team-BHP (https://www.team-bhp.com/forum/)

-   Shifting gears (https://www.team-bhp.com/forum/shifting-gears/)

-   -   The Generative AI Thread (chatbot, text to image, text to video etc) (https://www.team-bhp.com/forum/shifting-gears/259867-generative-ai-thread-chatbot-text-image-text-video-etc-9.html)

This is GPT4's (Bing AI Chat) response when I asked a question in Malayalam lol:



Of course, I set the language to Malayalam before asking.

Overall, the free version of ChatGPT is far better than Bing AI chat.

ChatGPT is a great coding assistant for a principal software engineer. Fresh college graduates and those who are new to coding may not find it that useful. You can create software within no time but with good code quality. You need to be familiar with object-oriented programming concepts, design patterns and you should have a good level of coding expertise. I am in love.

I have created a couple of projects entirely using ChatGPT. My value add was in giving ChatGPT the right inputs and refining its outputs. I even open-sourced one of that projects in Github. :D

Best to read up on how GDIs work. This is a complex long form answer involving significant engineering explanations. The short version - petrol and diesel behave differently.

Did some research and got the answer. Its not that complex to be honest.

The Self Ignition Temperature of Diesel is 210°C and that of Petrol varies from 247°C to 280°C.


Petrol: Less Viscous, more volatile, less flash point and more self ignition temperature.
Diesel: more viscous, less volatile, more flash point and less self ignition temperature.
Both flash point and self ignition temperatures should be considered to decide mode of ignition like spark ignition and compression ignition.

Also , Petrol lets its heat off with such a quick explosion that there isn’t enough time for it to heat the combustion chamber ready for the next cycle so it needs a spark plug to ignite it on each cycle.


Theoretically, a GDI engine with 1:22-28 compression will not require a spark plug, but it would be prone to uncontrolled explosions (knocking), and would be expensive to manufacture and maintain. Thus, the GDI engines dont go to that compression range; and hence get away by using spark plugs.

Now it makes perfect sense.

Bing AI search is just dumb compared to ChatGPT.

GPT-4 Saves Dog’s Life - https://analyticsindiamag.com/gpt-4-saves-dogs-life/

Now this is really cool. With the right safegaurds in place, these technologies can complement and revolutionize the healthcare sector and other burning areas.

As a country, we could definitely use some smart augmentation around health and education to deliver public welfare effectively. Right now the quality of both in the affordable category is abysmal to say the least and these generative models can help in lowering the costs, improving access and timely delivery atleast in a few non critical areas as a sratring point.

Really hope as humanity, we will get past the phase of gimmicky use-cases (chatgpt drafting emails, preparing travel itineraries) soon and leverage these fantastic tools for tangible societal impact whereit is needed the most.

Absolutely! I think remote consulting is already implemented in a few areas. The next step will surely be AI-driven consultations over a mobile phone for minor aliments or the initial consultation.

While not an AI use case, more of pattern-matching, was wondering if pattern-matching and image learning models could be applied to automate the routine aspects of lifeguard duties. Some beaches, like Goa's, have lifeguards who sit all day in the sun watching a mostly empty beach. Could there be cameras scanning portions of the beach and pattern-matching be used to identify a swimmer/kayaker in trouble in the water? With an alert sent to the tower for human intervention? (Of course on days where the beach is crowded that would involve human monitoring.) Any pattern-matching experts here, for informed comments?

This is practically a computer vision use-case which is a branch of AI :)

Theoretically yes it's possible. Like most AIML problems, this one will also depend on the quality and quanity of data available to train the model. Essentially it will be a classiifcation outcome where the model needs to look at every floating object (humans, animals, surf-boards, etc etc) and throw out whether the object is in danger or not in danger. If a large data set of such videos/images available, then it could be worth a try.

But practically, even before we even go towards an AI solution, a good first step could be just to have cameras installed on a beach and try monitoring by real humans remotely through a command center. Just like malls and other large areas have a CCTV room. That itself will save the lifegaurds from sitting in the sun.

Also if it turns out effective, then this solution can be replaced by AI on a larger scale. It also has the side benefit of generating real life data through the video feeds that can be annotated and used for model training later.

PS: I am not a pattern matching or AI expert but have some experience around business application of the technology.

Elon Musk, Apple co-founder Steve Wozniak and former presidential candidate Andrew Yang joined hundreds calling for a six-month pause on AI experiments in an open letter — or we could face "profound risks to society and humanity."

https://www.cbsnews.com/amp/news/elo...pen-letter-ai/

Meh. :ZZZ:

Let AI transcend humanity... Humanity isn't God's gift. If AI becomes supreme, so be it.

May be some other living organism is far beyond AI as well (Fermi' paradox). clap:

Wow, shades of past (usually religious) opposition to everything from Astronomy to Stem Cell research? Surprised to see two people who founded companies that have benefited directly from ground-breaking research now on the pause/ban side. Goes to show that everyone, even those who start off as anti-establishment, eventually end up becoming the establishment.

Agree. Let it happen. Someone may try to put a check on it, but it will definitely have its course of 'evolution' and its impact on humanity.

The rise of computers and the interconnected world is already slaying the almost outdated dogmas like religious beliefs and the belief in an all-seeing, all-controlling celestial entity (or entities), obsolete political ideologies, and anything else that has to be irrelevant to modern civilization. Despite this, the majority of the world remains entrenched in all sorts of obsolete ideas. May it will take AGI to continue this 'crusade' lol:

To avoid any potential backlash for labeling religion as "almost outdated," let me clarify that my time reference spans a vast timeline encompassing human evolution, the agricultural and industrial revolutions, the emergence of artificial intelligence, and far beyond.

A recent viral video:

"...the most intelligent inhabitants of that future world won't be men or monkeys, they'll be machines - the remote descendants of today's computers. Now the present day electronic brains are complete morons, but this will not be true in another generation. They will start to think and eventually they will completely out think their makers. Is this depressing? I don't see why it should be. We superseded the cro-magnon and neanderthal men and we presume we're an improvement. I think we should regard it as a privilege to be stepping stones to higher things." -Arthur C. Clark

https://www.youtube.com/watch?v=YwELr8ir9qM&t=259s

Got a summary of long threads of dashcam on team bhp.

For those interested in analyzing company balance sheets before investing.

https://www.youtube.com/watch?v=ApMMSacmwXI

Everyone knew this was coming, isn't it?

https://www.bbc.com/news/world-us-canada-65452940