Can AI and Robots do your job?

[Spuds]

Originally Posted by Irace86.2.0

I put it in and got this:

Oddly enough, that's about what I would expect to get from a coder lol ( ). But that's not really the answer to "What does this do?", it's simply describing what it contains and cross referencing it with statements from web pages in it's memory describing it. The chatbot does an ok job making it look like a human-written description, but if you remove the flourishes on either side of the root words, it becomes apparent it's just reading and regurgitating.

[Irace86.2.0]

Originally Posted by Spuds

Oddly enough, that's about what I would expect to get from a coder lol ( ). But that's not really the answer to "What does this do?", it's simply describing what it contains and cross referencing it with statements from web pages in it's memory describing it. The chatbot does an ok job making it look like a human-written description, but if you remove the flourishes on either side of the root words, it becomes apparent it's just reading and regurgitating.

It might be the case that I need to ask it a more specific question. What question would you want answered? The more vague, the more variable the interpretations.

Reading and regurgitating is exactly what we do. We may not be aware of what our subconscious is doing to extract “the file”. In fact, we are pretty bad about recognizing what we pick up and are regurgitating. I remember this Derren Brown show where multiple teams had to come up with a novel commercial/ad, and they all did the same exact idea. Why? Because everyone individually had been taken on a tour past triggering sites.

Try to think of a test that would differentiate an AI or Data on the show Star Trek from ChatGPT instead of something compared to a human, so you could know that it had reasoning or consciousness.

[Clipdat]

https://en.wikipedia.org/wiki/Turing_test

[Irace86.2.0]

Originally Posted by Clipdat

https://en.wikipedia.org/wiki/Turing_test

It probably isn’t a good test of intelligence, but of how similar a system conveys typical human speech patterns. Many humans would fail this test. Data from Star Trek would likely fail the test, but is clearly sentient.

Are we trying to determine if these systems will be the ultimate spam bot, be capable of understanding spam from troll on a car forum in order to be a moderator, or be intelligent enough to be sentient? I forgot.

GPT5 might be able to reach Artificial General Intelligence (AGI), which is the level of intelligence where an AI system can perform any task that a human can do. Some experts believe that GPT-5 could pass the Turing test, which is a test of whether a machine can exhibit human-like behavior in a conversation.

https://dataconomy.com/2023/04/chat-...ning-features/

[Irace86.2.0]

Latest version of ChatGPT aces bar exam with score nearing 90th percentile

https://www.abajournal.com/web/artic...0th-percentile

https://youtube.com/shorts/3x95mw35dJY?feature=share

[Clipdat]

Originally Posted by Irace86.2.0

Are we trying to determine if these systems will be the ultimate spam bot, be capable of understanding spam from troll on a car forum in order to be a moderator, or be intelligent enough to be sentient? I forgot.

"Is this testing whether I'm a replicant or a lesbian, Mr. Deckard?"

[dpfarr]

What’s a tortoise?

[Irace86.2.0]

This is pretty crazy stuff. I don't think the world is prepared for how disruptive these tools are. The results are not only crazy, but the speed at which they are advancing is even more crazy.

[NoHaveMSG]

Originally Posted by Irace86.2.0

This is pretty crazy stuff. I don't think the world is prepared for how disruptive these tools are. The results are not only crazy, but the speed at which they are advancing is even more crazy.

https://www.cbsnews.com/news/michael...rview-lawsuit/



Michael Schumacher's family is suing over an AI generated interview that was recently created. Curious what type of precedent this will set. Deepfakes have already been around for quite a while. Some of them damaging and in poor taste, some hilarious.

[Irace86.2.0]

My prediction is that within ten years, maybe five, entire industries will be seeing massive layoffs. Movie production is one place, artists of course, editors/writers, lawyers, coders, market analysts, CEOs/company operators/managers, remote workers, banks, OnlyFans, etc.

Everything in the world is moving so fast and will be so disruptive, I think people are going to really start freaking out.

[Clipdat]

On a scale of 1-10, how stoned are you right now?

[Spuds]

Originally Posted by Irace86.2.0

My prediction is that within ten years, maybe five, entire industries will be seeing massive layoffs. Movie production is one place, artists of course, editors/writers, lawyers, coders, market analysts, CEOs/company operators/managers, remote workers, banks, OnlyFans, etc.

Everything in the world is moving so fast and will be so disruptive, I think people are going to really start freaking out.

...nurses, doctors, technicians...

[Irace86.2.0]

Originally Posted by Clipdat

On a scale of 1-10, how stoned are you right now?

I don't smoke or drink. Watch the videos. The progression of the software in four years from making fuzzy pictures to photorealistic pictures is pretty remarkable. I think it'll be faster for video. The amount of users adding to this tech has drastically increased, which will only make it better.

Tell me why I am wrong?

[Irace86.2.0]

Originally Posted by Spuds

...nurses, doctors, technicians...

I would hope so. If we could decrease the cost of healthcare while increasing the access and improving patient outcomes at the same time then the sooner the better

Definitely doctors and advise nurses for telemedicine and for consulting specialists. Less for techs. I'm sure healthcare providers will be able to replace a lot of non-emergency hospital visits for sure. AI already has bested the average doctor on taking the boards and on diagnosing diseases. The biggest use could be for secretaries setting up appointments, data analytics for avoiding hospital associated infections or optimizing operations, medical records, billing, coding for software, IT, etc.

There are a few obstacles though. First off, the healthcare industry is often far behind on technology. This is to save costs by avoiding technology turnover costs and to maximize profits, but also because anything medical is expensive. Wires that should cost $20-50 are $500. The second reason is related, which is that everything has to be approved for healthcare, which makes everything expensive, but it means it needs to be stamped by government agencies ten times. Like Apple releasing their ECG app to detect a-fib, it had to go through clinical trials to be approved as a medical device, so they could make a claim about what it could do (1). Lastly, hospital systems would need to borrow this technology and pay for it, which is possible, but I'm sure it would require federal funding to streamline/reduce operating costs (Medicare/Medical/Medicaid), but that type of implementation can take time.

The industries that the videos talked about that would be the least affected were labor based jobs where someone needs to physically move something. Until we have robots to do these things, there needs to be a doctor to intubate, put in a central line, do a heart cath, and there needs to be a nurse to deliver medications, place a foley, put in an IV, hang some antibiotics, do a breathing treatment, or techs to stock, do CPR/place Lucas device, do ECGs, restrain psych patients, etc. The profession is very physical and requires a lot of assessments and interventions.

I could imagine AI and autonomous driving could reduce accidents on the roads, decrease the size of a number of different healthcare departments and could reduce traffic to hospitals with AI telemedicine.