Just 16% of Americans think AI will benefit society, despite chatbot use climbing to 49% of US adults

[rmcrys]

I think many see the AI totally wrong. The AI is advancing brutally fast ("thank" the US) which is good and bad:

- there is barely any legislation
- the AI is dangerous, very dangerous. Why? Because it is a ML that can either be a tool (help someone accelerate per simplify their jobs, casual chat, look for something, etc) but it can also be a internet bot that steals human and non human information and use it to end with people's jobs, fake photos and videos and if used in the wrong hands, extremist stuff.

So, the AI is extremely helpful but also extremely dangerous. The same as nuclear: great stuff for energy or defense against meteor/aliens etc..... But also very dangerous/ends of planet life events.

There should already exist very strict legislation for what and how far it can be used. Help people? Yes. Substitute people or end jobs? No

 

[Endymio]

My daughter just had an MRI last week. They had a cheaper option, as I recall for one third the price, to have an AI reading instead of the doctor. I told her HELL NO, pay the doctor, it was a potentially serious organ issue. Not something you want to put in the hands of AI, I don't care how good it's claimed to be. lol

Last year, I paid a radiologist $5K and waited almost 3 days to results -- he missed a problem he'd been instructed specifically to look for. The AI scan found it in seconds, and gave me the results that same day.

 

[TechSavy101]

This is prime example of a government being bought off by corporate and becoming complacent in corruption, conspiracy and crime. The result will be chaos. Its like the Italian mafia just bought off the mayor & entire New York police force. Nvidia is the Arasaka irl from Cyberpunk 2077.

 

[Steveb8189]

I use Alexa AI chat every single day since it's been activated in our house (a few months). It's FREE and it finally made the device useful more than than playing music and turning lights on an off. It's "dumb" AI, I cannot train it, it doesn't remember what we talked about after about 2-3 minutes of no interaction.

Honestly, it's one of the greatest inventions ever. I use it for information, data lookup (more for work), complex calculations and estimates. It's like having the Star Trek bridge computer at my disposal. After a few months I've trained myself how to use it most effectively and it's shockingly powerful when flexed to it's full potential. THAT SAID - It can be wrong on specific topics but, even more interesting, I can tell it to "look deeper and longer" and it will - sometimes correcting it's previous answer. I could write a full article about my experience with Amazon Alexa AI. It's that cool!

For our house, we'd be crippled without it. It's like having our own intelligent helper. We'd be back to fully using computers and cell phones to look up stuff. It's already saved us HOURS in manually searching and looking-up alone. It does all the hard work with a simple voice command.

Nobody else here is using it?

It's not FREE. It's included in your prime membership - which you pay for.

 

[Endymio]

This is prime example of a government being bought off by corporate and becoming complacent in corruption, conspiracy and crime. The result will be chaos. Its like the Italian mafia just bought off the mayor & entire New York police force. Nvidia is the Arasaka irl from Cyberpunk 2077.

This is so mind-bogglingly wrong-headed it makes me wince. Either the US keeps the lead in AI, or China takes it, at which point the US exports nothing but corn and pork bellies, while the world marches to China's tune, politically, economically, and militarily. Consider yourself lucky you're spouting CCP propaganda here, and not anti-Chinese messages there, or you'd be in a re-education camp before your fingers left the keyboard.

 

[gingerbill]

Last year, I paid a radiologist $5K and waited almost 3 days to results -- he missed a problem he'd been instructed specifically to look for. The AI scan found it in seconds, and gave me the results that same day.

lol , I am sure this actually happened.

 

[Endymio]

lol , I am sure this actually happened.

The nice thing about facts is that they remain true, no matter how you deny them. But my anecdotal evidence pales compared to the great number of medical studies showing AI already outperforming human radiologists. Here's just two of hundreds:

AI Outperforms Radiologists in Pancreatic Cancer Detection

International study finds AI detects pancreatic cancer on CT more accurately than radiologists, offering earlier diagnosis and improved survival prospects.

www.emjreviews.com

AI Algorithm Outperforms Radiologists in Detecting Breast Cancer | Docwire News

A retrospective study measured the diagnostic value of an artificial intelligence (AI) algorithm to detect breast cancer and found that it outperformed the detection ability ...

www.docwirenews.com

 

[ScottSoapbox]

I think you've kind of missed out on how the medical research community is using it. They're not cranking out new drugs and rushing them into market. They're actually using AI in a very good way. Rather than use the internet to search out answeres from crackpots, they're programming medical biological, and chemistry datbases, which the AI (ML?) take that information and suggests compounds and medicine that MIGHT work. And the research scientists do the heavy lifting of evaluating them through the normal proceedures.

It already has some promising drugs being tested now.

I think you missed out on how I was responding to the Pew survey of normal people in the article, not what scientists are doing with AI.

20% of normal people said they are using it for medical advice. AI research is great but in no way changes the reality of what I said.

 

[Squid Surprise]

I think you missed out on how I was responding to the Pew survey of normal people in the article, not what scientists are doing with AI.

20% of normal people said they are using it for medical advice. AI research is great but in no way changes the reality of what I said.

Provided you back up your search with real research and a doctor consultation (if it’s serious) there’s nothing wrong with consulting AI about your health.

My father is a doctor and he says the AI chatbots often give good advice - but they shouldn’t be trusted on their own… kind of like asking a friend about your aching tummy… no problem with it, just do a bit of research as well…

A doctor tends to have a big advantage because of physical exams, prior knowledge of the patient and access to medical tests (X-rays, ultrasound, MRI, etc) and drugs that a chatbot obviously doesn’t have.

Having another tool is never a bad thing though!

 

[Gars]

You are using it wrong

 

[Gars]

 

[yannus]

In the late 19th Century, most of the world public hated and feared electricity: newspaper and magazine articles referred to it as a incredibly dangerous "unrestrained demon", and prominent individuals declared they'd never, ever allow it into their homes.

They won't fear anymore as AI will take all the electricity. Being less sarcastic, most people in the tech industry see a great deal of danger, not from AI itself but from the corporations and governments owning them.

 

[yannus]

Last year, I paid a radiologist $5K and waited almost 3 days to results -- he missed a problem he'd been instructed specifically to look for. The AI scan found it in seconds, and gave me the results that same day.

5K?!!! That's no normal price.

 

[Jarenien]

Probably consists of the same mouth-breathers that thought DLSS was a waste of time and would hurt gaming. People are just ignorant/stupid when it comes to technology. AI is going to save the human race. do a !RemindMe 5 years and come back here and see that I am right.

 

[Gars]

Probably consists of the same mouth-breathers that thought DLSS was a waste of time and would hurt gaming. People are just ignorant/stupid when it comes to technology. AI is going to save the human race. do a !RemindMe 5 years and come back here and see that I am right.

you are talking for something godlike?
if ppl continues this routine of escalating problems, they don't deserve to be on this planet.
DLSS will come after this. Now its just fancy expensive tech. Game play is far more important, and here is the human power.

 

[Tantor]

My daughter just had an MRI last week. They had a cheaper option, as I recall for one third the price, to have an AI reading instead of the doctor. I told her HELL NO, pay the doctor, it was a potentially serious organ issue. Not something you want to put in the hands of AI, I don't care how good it's claimed to be. lol

That's a very interesting situation.

1) Why would the AI analysis cost anything at all? The cost to the hospital would be nominal. It should be free.
2) I would have done both analyses, and wanted to see them separately. The AI analysis would probably be far more in-depth. If the doctor's analysis didn't match the AI, I would ask him why his is better. If he made a compelling case that his analysis was better, I would ask why the hospital is offering inferior care...
3) Finally, I would ask if the doctor was using AI without telling me.

 

[abactuon]

The 20% using AI for medical advice is terrifying.

Although my medical doctor friends often lament Dr Google so maybe that’s not a new problem.

I'm curious what doctors say about the Global Medical Knowledge Base (MSDManauls.com).
Can I use it or, for example, check the qualifications of my treating physician?

 

[Endymio]

1) Why would the AI analysis cost anything at all? The cost to the hospital would be nominal. It should be free.

Eh? A larger hospital will pay in the neighborhood of $100K a year for a license for these models, and -- even were it free -- they must still recover tort liability costs on any service they offer. In today's US, even something as simple as a nurse dispensing a patient an aspirin can trigger a billion-dollar lawsuit, and that cost is born by consumers.

That said, the hospital charged me ~$60 for my bundle of scans ($5/ image) which was two orders of magnitude less than a human radiologist.

3) Finally, I would ask if the doctor was using AI without telling me.

Do you ask your doctor if he used any reference materials, medical journals, or any other tools to assist him? Under medical ethics, if a doctor believes an AI tool would improve his diagnosis, he's duty-bound to use it.

 

[ScottSoapbox]

I'm curious what doctors say about the Global Medical Knowledge Base (MSDManauls.com).
Can I use it or, for example, check the qualifications of my treating physician?

I'm not familiar with that site, but I can confidently say that many doctors were not at the top of their class or even at a good med school. (Getting my own doctorate made this abundantly clear)

For routine stuff it really doesn't matter, but for important things you should definitely know your doctors qualifications and seek out someone beyond who is assigned at the most geographically convenient location (which is usually how one picks a doc for routine stuff).