[tranquility]

Originally Posted by allinon72

Public facing AI like these chats are going to be pre-programmed, neutered versions which amount to propaganda machines, as we're already seeing. There's no incentive for the leaders of this world to release an all-knowing entity that will willingly unleash the truth and not abide by the accepted narrative at any point and time. I find it weird that these chat AI programs just showed up one day, multiple versions, in synchronization. We aren't in Kansas anymore, everything put in front of us has an agenda, and most of the time it isn't in our best interest.

They will keep the all-powerful AI behind the scenes, as they always do.

Don't bother trying, many ppl just refuse to see what's up, for example, Conan O'Brien did a skit where he showed clips from news anchors saying exactly the same thing yet ppl still think they are getting unfiltered agenda-free 'news'.

[ezaircon4jc]

Just who(m) programs the initial AI? It doesn't get its "knowledge" from thin air. AI will NEVER be able to adjust to the nuances of human interaction or emotions. We are infinitely more than ones & zeros. Granted, I don't follow the AI thing much.

[bimmer456]

You can use it to code apparently. Imagine making an iPhone app and just describing what you want. Also using it to make artwork or nft

[bimmer456]

Also integration with home automation assistance

[AmuroRay]

DARPA and the CIA had more advanced AI than this 20 years ago, this is just the consumer version of the product.

I started noticed AI botting during the previous election (Clinton/Trump) and it really came to head during COVID.

Reddit has a CHATGPSIM (version 2 or 3) that is basically a whole subreddit where the bots make topics and talk to eachother. It’s freaky how the algorithm picks up on the conversations all over social media and basically can mimic human like responses even down to ‘telling stories’

And this is just the consumer version - not what the government uses/employs

Combine this with Deepfake technology and you’ll likely never know when you’re talking to a person or a bot.

[AmuroRay]

Originally Posted by allinon72

Public facing AI like these chats are going to be pre-programmed, neutered versions which amount to propaganda machines, as we're already seeing. There's no incentive for the leaders of this world to release an all-knowing entity that will willingly unleash the truth and not abide by the accepted narrative at any point and time. I find it weird that these chat AI programs just showed up one day, multiple versions, in synchronization. We aren't in Kansas anymore, everything put in front of us has an agenda, and most of the time it isn't in our best interest.

They will keep the all-powerful AI behind the scenes, as they always do.

100%.
This is clearly DARPA/CIA, and will (and are) being used as propaganda tools. Instead of telling you to do your own research, CHATGP will give you the answer they want you to have.

[FunnyUsername]

Originally Posted by bayarea328xit

If you are interested in following the recent developments in AI, I think the following article is a good start: Microsoft Research: Early experiments with GPT-4 (from March 27 2023).

If these systems can routinely solve problems like what is shown below (from top of page 40 in the above article), we are on the verge of a big change in how and what we have to teach in schools and colleges - the impact on the workforce will be huge.

My workplace is actually discussing experimenting with Github Copilot in development pipelines. If this is successful and becomes widespread in industry, productivity will vastly increase, but SDE jobs will lose demand and could impact employment/wages.

[antzcrashing]

ChatGPT is dumb. It can’t answer basic solve for x math problems.

It will fail to impress even average intelligent humans, and its credibility will erode really quick.

Its main utility is to pump stocks like NVIDIA, microsoft, google etc





Flame suit on 🔥

[dradernh]

Originally Posted by antzcrashing

ChatGPT is dumb.

Its main utility is to pump stocks like NVIDIA, microsoft, google etc

I'm interested in learning how stocks worth from many hundreds of billions to a couple of trillion dollars can be pumped. Is that something you can explain?

[antzcrashing]

Originally Posted by dradernh

I'm interested in learning how stocks worth from many hundreds of billions to a couple of trillion dollars can be pumped. Is that something you can explain?

Lol, no I can’t explain that to you. Flame suit still on

[Chick Webb]

Well, I didn't expect much from an auto forum and have been rewarded appropriately.

Whether you believe it, yet, or not, generative AI using LLMs is a technology that will revolutionize many aspects of our lives. And change the way many of use work, if we still have jobs, as well. My team has been using ChatGPT (we call him Phred) to improve our productivity at work for a couple of months now. Several of the platforms that we use have already integrated LLM-driven content generation capabilities to expand and enhance their features. We are doing the same in our content management systems. I personally have used Phred to help me write some particularly tricky code, Excel spreadsheet formulas.

One example. We had a writer who happened to be British write a blog about the energy situation in the UK and what the government was doing about it. Now, maybe she's not a great writer, but the draft she submitted was written very much from the perspective of a Brit, and assumed the reader had a lot of background in the workings of UK government and society. Which didn't work for our audience. So I fed it to Phred with the prompt to rewrite it into American English and to assume a minimal knowledge of the UK. As fast as you can type it was done. And close to perfect; just needed a few edits.

Another thing that's very interesting is the ability of these models to translate into Asian (character-based) languages. In the past you really couldn't do this. Google Translate did not understand the nuances and generally did a shitty job on Chinese, Japanese, Korean, and the like. Not so with ChatGPT (esp GPT-4). We've been using Phred to translate a bunch of content into those languages and our in-country reviewers are only making very minor edits, mostly around jargon related to our industry niche. To address that part, we are working with a translation platform vendor to fine-tune GPT-4 using our translation libraries and when we do that will no longer have to pay translators to do this work.

I've been in tech for over 40 years and this is the third time I've seen such a transformational technology emerge. The first was computer networking in the late 70s, the second was the modern internet in '94. This is the latest and will likely have the greatest impact of any we've seen to date. Companies and individuals will either get on board with this technology or they will get run over. It's even created a new job category; prompt engineering. Seriously, Google it. Or ask Phred.

[SigmaShift]

I actually implemented GPT into a few apps that I've worked on as head of product. It is not a silver bullet, but it definitely can cut down on some tasks.

[Chick Webb]

Originally Posted by SigmaShift

I actually implemented GPT into a few apps that I've worked on as head of product. It is not a silver bullet, but it definitely can cut down on some tasks.

Yeah, definitely not a silver bullet, but already this technology is capable of making a marketing content creator or software development engineer significantly more productive, replacing a translator, etc. And it's just getting going. The models are now being trained to train themselves, which I believe is going to result in exponential growth of the capabilities of this technology. It's going to get very interesting.

[BMWGUYinCO]

So we recently created a new department within our IT organization: Application Delivery Services (think Citrix, RDS, etc.).
I was curious about using ChatGPT so I asked it to create a job description based on 3 tiers of experience with skill sets specific to that function. It produced perfect job descriptions almost instantly.
Saved me some time and research for sure.

[BMWGUYinCO]

And speaking of AI in the news....(spoiler alert: potential "Get Off My Lawn" rant coming) a lot of the music today kinda sucks. It all sounds the same in some genres.

To prove the point, some clever folks are using AI to generate entirely new songs from the likes of Drake and others - it's that easy. Now, try that with a Van Halen song.

[wtwo3]

[AmuroRay]

Originally Posted by wtwo3

Attachment 3160399

Attachment 3160400

😂😂😂😂

If they are known, they aren’t safe

[omnimous]

Long video but some scary points not many bring up

[Chick Webb]

Originally Posted by omnimous

Long video but some scary points not many bring up

The overriding theme of that video, for those that didn't watch, is that things are moving very quickly and we don't fully understand where they're going. The presenters (from the Center for Humane Techology) argue for slowing the progress of these Gollems (GLLMs) out of an abundance of caution.

I don't necessarily disagree that the pace of progress is concerning - as I noted in a post above, "The models are now being trained to train themselves, which I believe is going to result in exponential growth of the capabilities of this technology.". Given the opportunities to gain power/wealth through their development/deployment I think it's going to take a lot more than "concern" to put the brakes on this, though. Likely something bad will have to happen before enough people get motivated to do something about it. As always.

[DieselDiner]

The 'Godfather of AI' says we need to slow down:

Geoffrey Hinton was an artificial intelligence pioneer. In 2012, Dr. Hinton and two of his graduate students at the University of Toronto created technology that became the intellectual foundation for the A.I. systems that the tech industry’s biggest companies believe is a key to their future.

Dr. Hinton said he has quit his job at Google, where he has worked for more than decade and became one of the most respected voices in the field, so he can freely speak out about the risks of A.I. A part of him, he said, now regrets his life’s work.

More here: https://dnyuz.com/2023/05/01/the-god...-danger-ahead/

[2008M36MT]

I don’t know if it’s the same ChatGPT but I downloaded the iPhone app and can’t say that it’s super impressive. Maybe I’m asking the wrong stuff. Seems like Wikipedia was turned into an app.

[br438]

1) There is no stopping it. "Good gals/guys" can stop using it, but... the "Bad guys/gals" won't stop. so... do you want equal AI in your corner or not? It's the new nuke.

2) As we debate if/how/when it will mean the destruction of life as we know it, most miss the point that... the more anyone uses it, the faster the cycle becomes. We're all just trying to get ahead in life and hardly anyone ever frames their actions at work/play for moral responsibility... whatever that is.... I see net zero restraint of self edification at the cost of humanity across the world. ... and no. I don't know how to fix that... nor do I think I might figure it out. Maybe someone can pose the question to ... you know.

alas...

3) We (humans) are collectively a moron.