Predictions about the impact of AI-generated imagery

[ahalpert]

This is odd. Apparently these chatbots just make **** up sometimes.

“What it gives you is kind of like an Aaron Sorkin movie,” he said. Mr. Sorkin wrote “The Social Network,” a movie often criticized for stretching the truth about the origin of Facebook. “Parts of it will be true, and parts will not be true.”
He recently asked both LaMDA and ChatGPT to chat with him as if it were Mark Twain. When he asked LaMDA, it soon described a meeting between Twain and Levi Strauss, and said the writer had worked for the bluejeans mogul while living in San Francisco in the mid-1800s. It seemed true. But it was not. Twain and Strauss lived in San Francisco at the same time, but they never worked together.

Scientists call that problem “hallucination.” Much like a good storyteller, chat bots have a way of taking what they have learned and reshaping it into something new — with no regard for whether it is true.

https://www.nytimes.com/2022/12/10/t...t-chatgpt.html

[NorBro]

Uh, oh...first signs of having a mind of its own.

We're done.

[hojomo]

This is odd. Apparently these chatbots just make **** up sometimes.



https://www.nytimes.com/2022/12/10/t...t-chatgpt.html

Factions of scum***** the world over funding propaganda & misinformation campaigns must be particularly gleeful at the prospect of having dirt-cheap robots invent ever-more horrible distortions than their most cynical human actors. These systems have a way of leveraging the work of so many people and funneling it to serve existing power structures so well -- weird fluke of the technology. Total coincidence that it's all unraveled out to the public through unregulated 'non-profit' entities (in the west at least).

[DLD]

• Google has issued a "code red" over the rise of the AI bot ChatGPT, The New York Times reported.
• CEO Sundar Pichai redirected some teams to focus on building out AI products, the report said.
• The move comes as talks abound over whether ChatGPT could one day replace Google's search engine.

https://www.yahoo.com/finance/news/g...190131705.html

[NorBro]

uh-oh... that's a wrap, we're done

[DLD]

Since we're chatting about ChatGPT, someone posted this on Facebook. (I don't know who it is .. maybe someone in a sphere of influence of an acquaintance ... maybe someone even more random)

-----------------------------------------------------------

Cheating and Chat GPT


Yesterday I discovered that 8 out of 70 of my students had generated either parts of their terms papers or parts of their take-home Finals using Chat GPT; I've caught another 2 (update: 4!) this morning.

This is a far higher percentage of students cheating that I've encountered before at The College of (name withheld by me, DLD)

But after playing around with Chat GPT for a while I'm no longer worried at this--in large part because (obviously!) it's very easy to detect work that has been generated by it.

The work that Chat GPT generates is pretty distinctive. If asked about a controversial topic it will produce an anodyne paper along the lines of "Some say this, some say that, it's complicated". If asked to recount a theoretical point it will provide a short summary in neat declarative sentences. After playing with it for a while answers that have been generated by it become fairly easy to spot.

Of course, merely suspecting that a student's paper has been generated by an AI isn't enough to establish that they've cheated.

Luckily (well, for me...) it's very easy to use Chat GPT to do this.

I started off by typing in both my actual essay prompts and variants on them. Chat GPT doesn't (yet) vary its answers, and so this was sufficient to identify several cheating students. (This was made easier by students not realizing that others are using Chat GPT to cheat, and so handing in identical work!)

Other students were sneakier, and used Chat GPT's answer as the "framework" for their response. They then simply tweaked it themselves, or ran it through something like QuillBot, or moved around sentences or paragraphs with minor changes. And others had clearly typed in a slightly different prompt than the ones I'd used, and so received a different answer.

But these approaches to cheating were also simple to catch: I simply pasted their responses verbatim into Chat GPT. It then responded with something like "Yes, that's correct.... " and then offered its own response--which was that which the student had stolen and tweaked.

Reports of Chat GPT enabling students to cheat with impunity are greatly exaggerated.

As, alas, many of my students will be discovering next week.

With two of my five classes graded (70 students out of 175) I'm curious how many students will end up using this.

Then someone posted this in reply

It's on NPR, so have a ball.

A college student made an app to detect AI-written text : NPR

[DLD]

No, really, I was not speeding. And could not tell she was underage.

[DLD]

Nick Cave's opinion on ChatGPT songwriting ability - no bueno.

Nick Cave - The Red Hand Files - Issue #218 - I asked Chat GPT to write a song in the style of Nick Cave and this is what it produced. What do you think? Verse 1: In the depths of the night, I hear a call A voice that echoes, through the hall It’s a siren’s song, that pulls me in Takes me to a place, where I can’t begin Chorus: I am the sinner, I am the saint I am the darkness, I am the light I am the hunter, I am the prey I am the devil, I am the savior Verse 2: I walk the streets, with my head held high I’ve got my demons, but I’ll make them fly I’ll dance with the devil, and I’ll play his game I’ll be the one, to call his name Chorus: I am the sinner, I am the saint I am the darkness, I am the light I am the hunter, I am the prey I am the devil, I am the savior Bridge: I’ve got the blood of angels, on my hands I’ve got the fire of hell, in my eyes I’m the king of the abyss, I’m the ruler of the dark I’m the one that they fear, in the shadows they hark Chorus: I am the sinner, I am the saint I am the darkness, I am the light I am the hunter, I am the prey I am the devil, I am the savior Outro: So come with me, to the edge of the night Where the darkness and the light collide We’ll walk the line, between the good and the bad And we’ll see, who comes out on top in the end. : The Red Hand Files

[DLD]

Someone asked Chat GPT about (the US and Chelsea soccer player) Christian Pulisic. I am stealing the reply.

"It is difficult to predict how well Christian Pulisic will perform for Chelsea as it depends on various factors such as form, injuries, and the performance of the team as a whole. However, based on his past performances and his reputation as a talented player, it is likely that he will continue to be a valuable asset for Chelsea. In the past season he showed his quality and his importance for the team, scoring important goals and showing his skills."

Meanwhile Chelsea acquired $200M worth of players last week alone. Todd Boehly has deep pockets.

[ahalpert]

Someone asked Chat GPT about (the US and Chelsea soccer player) Christian Pulisic. I am stealing the reply.



Meanwhile Chelsea acquired $200M worth of players last week alone. Todd Boehly has deep pockets.

I have no idea who or what any of those people are. But that reply is pretty typical of chatgpt - not incorrect but not terribly useful or insightful.

I think once the system gets plugged into current internet search results, it has the potential to be more useful for sifting through data and presenting information in a concise way. The problem is that it's not always accurate, so in any circumstance where error can't be tolerated, I would still want to process my own Google search results or at least double-check the conclusions of the bot

[stoneinapond]

I have no idea who or what any of those people are....

Obviously you had no interest in the World Cup. Pulisic is one of your countrymen, who almost single-handedly tried to advance the USA national team to the next round of the competition. His valiant effort failed.