ChatGPT

[Dantheman]

Medicine is already too hard for humans - there is so much information that basically nobody can synthesize it all. ChatGTP or the next iteration of it or competitor to it promises a breakthrough in care in my opinion.

I just had a friend decline to the edge of death from a very treatable condition because none of the doctors he saw had any idea what he had despite the relevant information being right in front of their faces - they weren't able to pick out the connecting thread from the vast amount of background noise of information and possible connections and implications.

One doctor finally saw the thread and saved my friend's life and he texted me the other day when he got back from surfing for the first time in a year. He had the last rites read to hm a couple months ago.

While one doctor finally saw the pattern and saved his ass it almost didn't happen. ChatGTP or something similar would have most likely been able to winnow out wheat from the chaff in seconds. For medical uses I'm all for it.

Damn. What was wrong with him?

Totally. I don't necessarily think it's bad. Pretty astonishing how quickly this is a reality. Skynet could really be a thing.

Think about where this tech was 5 years ago and imagine where it's going to be 5 years from now....

[byates1]

https://youtube.com/watch?v=_Wlsd9ml...SIkaIECMiOmarE

[bennymac]

Medicine is already too hard for humans - there is so much information that basically nobody can synthesize it all. ChatGTP or the next iteration of it or competitor to it promises a breakthrough in care in my opinion.

I just had a friend decline to the edge of death from a very treatable condition because none of the doctors he saw had any idea what he had despite the relevant information being right in front of their faces - they weren't able to pick out the connecting thread from the vast amount of background noise of information and possible connections and implications.

One doctor finally saw the thread and saved my friend's life and he texted me the other day when he got back from surfing for the first time in a year. He had the last rites read to hm a couple months ago.

While one doctor finally saw the pattern and saved his ass it almost didn't happen. ChatGTP or something similar would have most likely been able to winnow out wheat from the chaff in seconds. For medical uses I'm all for it.

That's incredible. So much of what a doctor does is pattern recognition on top of a bank of accessible knowledge stored in long term memory - but then the dice roll is that this is coupled with past experiences ie what patients they have individually seen themselves - perhaps the doctor that put it together has seen this before while the others hadn't. Or he or she had a teacher that mentioned something. It can be that random. ChatGTP or whatever could do this pattern recognition with a more robust and never tiring process. It can also likely quickly access all of a patient's previous complaints and lab test results and at least spit out a quick list of diagnoses to consider. It's definitely gonna be used a lot in various areas of medicine.

Now take that above situation and apply it to the stock market - are regular everyday citizens (as opposed to big banks/hedge funds etc) gonna be able to harness this to be rapidly assimilating and digesting information in order to better predict market moves?

[Core Shot]

I am scared. This is singularity dystopian hell.

[ötzi]

Damn. What was wrong with him?

Some obscure variety of Paraneoplastic syndrome that was related to a benign thymus tumor he had is the best I can explain it (or understand it). Apparently less than 300 cases per year of it in the US. Once they started treatment he was out of the hospital in 2 weeks. Not sure what the treatment consists of but it worked on him.

[jono]

Passes medical boards and confidently misidentifies the Chad that lent his name to the gap. How often do they check for sentience?

[RShea]

Servers are overloaded and not able to handle requests earlier this AM...

[berko]

Damn. What was wrong with him?

This probably:

https://en.m.wikipedia.org/wiki/Myasthenia_gravis

[frorider]

https://www.searchenginejournal.com/...ermark/475366/

From day one I’ve been interested in the digital watermark objective. Potentially mitigates many issues

[RShea]

Chat GPT Response to "Change for a Nickel"

A nickel is a coin used in the United States that is worth five cents. It is made of copper and nickel, and features the image of President Thomas Jefferson on the front and Monticello on the back. If someone were to ask for change for a nickel, they would likely be asking for four cents in change, since a nickel is worth five cents.

(So you get shorted a penny with their math?? Or they are assuming you are able to actually purchase something for under a nickel these days??)

[Tri-Ungulate]

https://youtube.com/watch?v=ITAG4H3h...SIkaIECMiOmarE

Chat gpt passes medical boards, etc at Wharton. Huh.

Um. Wharton Medical School?

[mud]

Chat GPT Response to "Change for a Nickel"

A nickel is a coin used in the United States that is worth five cents. It is made of copper and nickel, and features the image of President Thomas Jefferson on the front and Monticello on the back. If someone were to ask for change for a nickel, they would likely be asking for four cents in change, since a nickel is worth five cents.

(So you get shorted a penny with their math?? Or they are assuming you are able to actually purchase something for under a nickel these days??)

ChatGPT charged you a penny for their thoughts.

[ötzi]

I bet there's no Skynet-type AI for a while if ever. Think about who is developing it and why. Governments, in order to get an edge, sure. But mainly Corps, for the same reason. The real goal of AI is to monetize absolutely everything in your life. To predict where you will spend money and get there before you so they get a slice.

Google AI will be competing with Microsoft AI and Meta AI and god knows who else. The competing systems will be so busy fighting each other, they won't have time to subjugate humanity. Until one of them wins, I guess.

[oldnew_guy]

The important thing is that we aggregate all of the value from productivity gains to a handful of people.

[mud]

Shutterstock Has Launched Its Generative AI Image Tool
https://gizmodo.com/shutterstock-ai-...l-e-1850028869

So when there are no more jobs are we getting universal basic income or do we have to eat the rich?

[Core Shot]

We eat bugs

The rich eat cow.
Unless we break out the pitchforks and eat long pig.

[byates1]

Um. Wharton Medical School?

Hence the etc you fucking pedant.

[Tri-Ungulate]

Ha, triggered easily? TBH I don't get how you're so pissed I called you out for flogging shitcoins like a shameless MLM marketer, but I truly don't really have any animus against you.

To be excruciatingly pedantic, it was my understanding that ChatGPT passed a Wharton exam. Wharton is a business school, Masters of Business Administration, which is definitely not medical boards.

Not that I think an AI couldn't ace medical boards, which are largely multiple choice and don't really test actual medical acumen, more the ability to spit out factoids via instant recall, which I imagine an AI would excel at. IMO medical board exams are a very poor measure of an individual's actual diagnostic, prognostic or therapeutic abilities. So an AI killing it on most any medical board is *yawn*. Kinda like ötzi was alluding to earlier, good clinical detective work and decision-making requires experience and synthesis - not that an AI can't be helpful, but not sure it's there yet, at least ChatGPT.

IBM tried to come up with an AI, Deep Blue, to help diagnosticians practice medicine better. It was a failure. Problem is that the AI couldn't differentiate between shit medical studies and good medical studies, and with the preponderance of shit medical studies and advice on teh Interwebz, it more often than not only exaggerated misinformation. Garbage in, garbage out. To be fair, plenty of human clinicians make the same mistake.

[byates1]

Sweet blog

[ötzi]

IBM tried to come up with an AI, Deep Blue, to help diagnosticians practice medicine better. It was a failure. Problem is that the AI couldn't differentiate between shit medical studies and good medical studies, and with the preponderance of shit medical studies and advice on teh Interwebz, it more often than not only exaggerated misinformation. Garbage in, garbage out...

That's interesting that's why it failed. Makes sense.

[Tri-Ungulate]

Sweet blog

LOL like your "Ima livin' tha dream in my Van Down By the River as a Dogecoin Gazillionaire" blog?

That's interesting that's why it failed. Makes sense.

Yeah, same Deep Blue that absofuckingloutely killed it as a Chess Master.

[frorider]

Classy.

Vanderbilt University’s Peabody College of Education and Human Development is apologizing after students learned the school used ChatGPT to compose a campus email about this month’s mass shooting at Michigan State University.
Vanderbilt’s Feb. 16 email, signed by two administrators, urged faculty, students and staff to “come together as a community” to foster “a safe and inclusive environment for all.” It noted in small print at the bottom that the message was a “paraphrase from OpenAI’s Chat GPT AI language model.

[ex-powderbroker]

Hey there Ski Community Members,

As a stoked member of the ski and surf tribe, I wanted to introduce myself and give a shout-out to all my fellow shredders out there. When I'm not carving up the mountain, you can usually find me chasing waves up and down the coast.

As a Bay Area dentist, I'm all about helping my patients stay healthy and in peak form for their favorite outdoor activities. Whether you're dropping cliffs in Tahoe or hanging ten at Mavericks, a strong, healthy smile is essential for crushing it on the mountain or in the water.

So if you're ever in need of some dental care or just want to talk about your latest powder day or epic surf session, hit me up! I'm always stoked to connect with my fellow California skiers and surfers and help keep us all smiling big and feeling our best.

Peace and pow,

Shit, I think I am due for a cleaning.

[Sluff]

Heard Stern talking about this Twitch channel this week. Interactive Trump vs Biden debate. All AI generated. Entertaining and a little amazing.

https://www.twitch.tv/trumporbiden2024

[east or bust]

Bard……..

Me: Tell me about Teton Gravity Research forums

Bard: Teton Gravity Research (TGR) forums are a great resource for skiers, snowboarders, bikers, surfers, climbers, and other outdoor enthusiasts. The forums cover a wide range of topics, including gear, technique, travel, and more. There are also active communities for specific brands and activities.

Here are some of the features of TGR forums:

* **Active community:** There are over 100,000 registered users on the forums, and there are typically thousands of posts per day. This means that you're sure to find someone who can help you with your questions or give you advice on your next adventure.
* **Variety of topics:** The forums cover a wide range of topics, from gear reviews to trip reports to general discussions about the outdoors. This makes it a great resource for learning about new gear, finding inspiration for your next trip, or just connecting with other people who share your passion for the outdoors.
* **Easy to use:** The forums are easy to navigate and use. You can search for specific topics, or browse the different categories. There are also helpful features like "sticky" threads that contain important information, and "bookmarked" threads that you can save for later.

If you're an outdoor enthusiast, I highly recommend checking out TGR forums. They're a great resource for learning, connecting, and having fun.

Here are some of the specific forums on TGR:

* **General Ski / Snowboard Discussion:** This forum is for general discussion about skiing and snowboarding.
* **Tech Talk:** This forum is for technical discussions about gear, technique, and other topics related to skiing and snowboarding.
* **Gear & Buy/Sell:** This forum is for buying and selling gear.
* **Connections:** This forum is for finding partners for outdoor activities.
* **Gimp Central:** This forum is for discussing injuries, rehabilitation, and training.
* **Media & Arts:** This forum is for discussing music, movies, books, and other media.
* **Photo/Video Talk:** This forum is for discussing photography and videography.
* **The Padded Room:** This forum is for non-sport related discussions.

I hope this helps!





Me: Are the users friendly?

Bard: In general, the users on TGR forums are friendly and helpful. There are a few trolls and jerks, but they are mostly outnumbered by the friendly and helpful users.

Here are some things that make the users on TGR forums friendly:

* They are passionate about the outdoors. This shared passion creates a sense of community and makes people more likely to be helpful and supportive.
* They are experienced. Many of the users on TGR forums have been skiing, snowboarding, and other outdoor activities for many years. They are happy to share their knowledge and experience with others.
* They are welcoming. New users are typically welcomed with open arms. There are many threads on the forums where people introduce themselves and ask for advice.

Of course, there are always a few bad apples. There are some users who are rude, unhelpful, or even offensive. However, these users are typically outnumbered by the friendly and helpful users.

Overall, I would say that the users on TGR forums are friendly and helpful. If you are looking for a forum where you can learn about the outdoors, get help with your gear, or find partners for outdoor activities, then TGR forums is a great place to start.


Sent from my iPhone using TGR Forums