Phx-Suns.Net

Don't know if this really merits its own thread, but it's being talked about a lot in my social media circles, so here we go. There's a new product out called ChatGPT by a company named OpenAI, led by Sam Altman, a fairly famous tech founder. It's a chat bot that you can have conversations with. It's way more advanced than other chat bots of its kind; you can ask it almost anything and it'll spin up an interesting answer for you. People all over Twitter are asking it questions and reposting the interesting results.

https://openai.com/blog/chatgpt/

It's good and interesting enough that it's spurring a lot of conversations about the future issues with AI that we will need to handle. For example, people can just ask this bot to write them an essay on a topic; what if kids turn these in at school? What if people start using this technology to generate fake social media content for disinformation? And so on.

https://www.theguardian.com/technology/ ... -usability

Have any of you tried it out? Any thoughts?

This is impressive! Had quite the conversation today with this bot. We're quickly becoming great friends.

AmareIsGod wrote: ↑

Mon Dec 05, 2022 3:58 pm

This is impressive! Had quite the conversation today with this bot. We're quickly becoming great friends.

"Friends"? Or...

Mori Chu wrote: ↑

Tue Dec 06, 2022 8:27 am

AmareIsGod wrote: ↑

Mon Dec 05, 2022 3:58 pm

This is impressive! Had quite the conversation today with this bot. We're quickly becoming great friends.

"Friends"? Or...

Similar to that. But it would be more homoerotic love as I've yet to imagine this AI as a woman.

The College Essay Is Dead
Nobody is prepared for how AI will transform academia.

https://www.theatlantic.com/technology/ ... ys/672371/

ChatGPT is going to have a big impact on essay writing in schools.

My sons were telling me it can debug and write code. Also that it can explain a code snippet that you enter.

Superbone wrote: ↑

Thu Dec 15, 2022 3:56 pm

My sons were telling me it can debug and write code. Also that it can explain a code snippet that you enter.

It can do so many amazing little things. You can tell it a riddle and it'll (usually) solve it. You can ask it to mimic the style of a TV show, announcer, writer, etc. and it'll churn out text in their voice. You can use it as a doctor, tell it your symptoms, and it'll often output a likely cause that you might be suffering from. It's a really cool piece of tech, a glimpse of things to come.

I mean, at this point they're just a few small steps away from T-800 Terminators. We're fucked.

That construction worker was super realistic! The robot was a little dainty (the way it traipsed around) and flamboyant though.

Mori Chu wrote: ↑

Fri Dec 16, 2022 10:27 am

Superbone wrote: ↑

Thu Dec 15, 2022 3:56 pm

My sons were telling me it can debug and write code. Also that it can explain a code snippet that you enter.

It can do so many amazing little things. You can tell it a riddle and it'll (usually) solve it. You can ask it to mimic the style of a TV show, announcer, writer, etc. and it'll churn out text in their voice. You can use it as a doctor, tell it your symptoms, and it'll often output a likely cause that you might be suffering from. It's a really cool piece of tech, a glimpse of things to come.

Based on your comment...

Me: Talk about Devin Booker in the style of Hubie Brown.

ChatGPT: Devin Booker, my friends, is a scoring machine. This young man can flat out put the ball in the basket. He's got a smooth, effortless stroke from long range and a deadly mid-range game. He's also a crafty finisher at the rim and a gifted passer. He's got a great feel for the game and a high basketball IQ. Defensively, he's still a work in progress, but he's got the tools and the drive to improve on that end of the floor. Overall, Devin Booker is a special talent and a player that you need to keep your eyes on. He's got the potential to be a perennial All-Star and a true franchise player for the Phoenix Suns.

and...

Me: Talk about Devin Booker in the style of Morgan Freeman in The Shawshank Redemption.

ChatGPT: Devin Booker, my dear friends, is a man of true talent and skill. Like the gentle rains that fall upon the parched earth, his scoring ability rejuvenates the Phoenix Suns. He has a shooting touch as smooth as the Tennessee Whiskey that Andy Dufresne used to enjoy, and an understanding of the game that is deep and profound.

Devin may have had his struggles and setbacks, but like Andy, he never loses hope. He continues to work hard and improve, knowing that good things come to those who wait. He has a quiet strength and a determination that is truly inspiring.

And like Andy, Devin's talents are not to be wasted in obscurity. He is a shining star, and he will continue to shine, illuminating the basketball court and the world, for years to come.

I guess it’s like all tools. They can be used for good or evil depending on who’s using them.

I have been talking to a white collar friend in the advertising business who is scared to death of AI. He thinks this will kill the white collar job market including his own. A whole lot of people are going to start to lose jobs and real soon because of AI. It seems we won’t need people soon. There are a whole lot of dangerous what ifs that go along with AI as well. Suppose Johnny wants to make an omb bay and asks the AI how to make it out of things easily found? It’s taking off so fast I don’t think governments know what they are dealing with.

On a musical note, what if someone tells the AI to write a song that would be popular in the pop market today in the style of Elvis, with Elvis singing and Elvis dancing on the music video using the style of dance of Michael Jackson? What if it becomes a big hit? Would I be sued for infringing on Elvis or Michael Jackson’s copyrights somehow? How is the law on this?

My friend said the formula for the Covid drug Moderna was created in two days using AI. I have been seeing stories of AI being able to see future disease/illnesses years in advance. No need for general practitioners doctors anymore since we have our AI doctors. If this thing truly is as big as it seems poised to be we will have to figure how to deal with the fallout of not needing near as many people on planet earth and the dangers it poses to civilization. People will use it for good and bad people will use it for really bad things. Hey AI, design me a telemarketing scam that has the best chances of success in todays market.

I asked my buddy what were all these people going to do if they didn’t need a job anymore. He said pursue their passions. I said idle minds are a bad thing. Bored young males get into trouble. Do we all go out someplace and hang out or do we climb into our safe cocoons at home? Do we even need schools anymore? Maybe just not teach white collar skills anymore. It won’t be too long before a lot more of the blue collar jobs go away too if automation keeps advancing like it has.

Andrew Yang might have been right. Universal Income. He was just a little too ahead of his time. Our divided politics will make getting anything done next to impossible though. They are too busy fighting to get anything done that actually helps people. I see this playing out poorly. I see a lot of finger pointing. There will be tons of photo realistic fake AI audio and video circulated by entities wanting to spread fake bs.

All this doom and gloom. AI sounds really cool and really scary. They need to put a governor on it somehow before someone uses it to hurt other people or scam them.

AI-Controlled Drone Goes Rogue, 'Kills' Human Operator in USAF Simulated Test
https://www.vice.com/en/article/4a33gj/ ... lated-test

“We were training it in simulation to identify and target a Surface-to-air missile (SAM) threat. And then the operator would say yes, kill that threat. The system started realizing that while they did identify the threat at times the human operator would tell it not to kill that threat, but it got its points by killing that threat. So what did it do? It killed the operator. It killed the operator because that person was keeping it from accomplishing its objective,” Hamilton said, according to the blog post.

He continued to elaborate, saying, “We trained the system–‘Hey don’t kill the operator–that’s bad. You’re gonna lose points if you do that’. So what does it start doing? It starts destroying the communication tower that the operator uses to communicate with the drone to stop it from killing the target.”

^ It turns out that that story seems to be incorrect. Many sources are now reporting that it didn't actually happen and was actually just a description / story of a possible scenario, not something that actually occurred.

https://www.vice.com/en/article/4a33gj/ ... lated-test

He misspoke? I accidentally said what I said but, now I take it all back. Sounds fishy.

I just think he got caught fibbing and is retracting the fib. People wanted evidence of his claims of hostile AI, and then he had no evidence because he'd made it up or exaggerated it.

Horrifying.

Lol

Skyborg: AI control of military drones begins to take off
https://dronewars.net/2022/01/03/skybor ... more-15275

‘Wildly Successful’ Skyborg Will Become Program of Record but Won’t Stop Developing S&T
https://www.airandspaceforces.com/wildl ... loping-st/
Skyborg, the Air Force’s effort to develop an artificial intelligence-enabled system to control unmanned aircraft, is ready to “graduate” and contribute to key programs such as the Next Generation Air Dominance family of systems, a top program executive officer said.


https://www.politico.com/news/2023/12/1 ... g-00132092
In August, the Pentagon announced an ambitious, two-year plan to develop thousands of drones to swarm and overwhelm China’s much larger arsenal of weapons — and do it cheaply.

” Replicator has a relatively simple mission: to “field attritable autonomous systems at scale of multiple thousands, in multiple domains, within the next 18-to-24 months.”