An Artificial Intelligence Thread

This is coming at us fast, and that won’t stop. Post here articles, thoughts, experiences, etc. Here’s a start:

I, for one, welcome our new overlords.


It seems the only logical place to start here is with what ChatGPT v3.5 has to say about it:

The future with increased use of artificial intelligence (AI) is likely to involve advancements in various fields such as automation, healthcare, transportation, and more. AI may lead to improved efficiency, personalized experiences, and innovative solutions. However, challenges like job displacement and ethical concerns will also need to be addressed. It’s a complex landscape with both exciting opportunities and potential drawbacks.


And here’s what it says about generative AI:

Generative AI, a subset of artificial intelligence, holds the potential to create content, images, and even entire scenarios. With further development, we can expect generative AI to revolutionize creative fields like art, music, and writing, enabling the generation of new and unique content. However, it also raises questions about authenticity, intellectual property, and the potential for misuse. As generative AI advances, its impact on various industries and creative processes is likely to be profound.


“AI” right now means the generative stuff, large language models, etc. But AI has been around awhile. We use machine learning a LOT, mining DNA data for patterns, looking for connections on DNA mutations, how RNA is expressed and what that means for different treatments.

I use GPT and Bing every day, for one thing or several.

Don’t believe everything you get in Google search results, and don’t accept code or answers from the GPT family of LLM without further validation. That said, it helps accelerate a lot of my work quite a bit. It’s a little like the skills acquired over the years in becoming a “Google whisperer”.

LLM’s are powerful but also limited, and just as the City of San Francisco just approved fully autonomous taxi rides from Waymo and Cruise, there’s no way to hold back technology, but managing the transition to the future is absolutely essential.

(I’ve been studying a bit about the economic history of the 1920s. Wow. I don’t know how our nation got through that period. It was absolutely brutal. For example, in one year 600,000 farmers lost their farms, most moved in with relatives or came to the cities in search of work. The related resentment of immigrants and blacks coming out of the South was vicious.)


Yep. If I dictate a note or email to my iPhone or take advantage of autofill, I’m using AI.

1 Like

I have a friend, a young rising star TV writer and member of the Writers’ Guild of America (currently on picket lines) who’s very worried about this. He’s just starting his career and seemed to be on the path to success.

I think there was a recent court case where the judge rules that you couldn’t copyright AI made content. From my extremely limited viewpoint, that seems like a big win for writers.


How do you ascertain when content was made by AI or from an individual to verify the copyright?

What about those who offer stock or other advice? They can generate company reports and news with AI and put it up on their “pay for” sites. Would it be legal for someone to capture the AI generated content from the paid site and then publish it on a free (or even another paid site)? What recourse does the original site have?

1 Like

I also have a friend who’s got both writer and actor guild cards and is currently in LA picketing. He’s one of those side character types, he was on the “Ghostbusters” reboot with the women heros as the ghost in the Subway and on “Frazier” as a driving instructor, so he’s got some serious concerns with image use.


Don’t blame him. I can see that being abused across the board, from games to tv or movies.


Isn’t a lot of the problem that AI can’t create unique content? That it’s just a computer that scans the internet and copies and pastes the best it can to give you what you want? If that’s the case…they how can you copyright that? If anything, wouldn’t AI leave you open to lawsuits for not properly sourcing your material?

Someone here mentioned the federal court decision that this blog post discusses. This is a trial court decision. I hope it’s upheld on appeal. Early in my life as a lawyer I did a lot of copyright work, and the court’s opinion seems quite sound legally.


Me too.

Axios excerpt:

Websites revolt against AI giants

Nearly 20% of the world’s top 1,000 websites are blocking crawler bots that gather data for AI services, Axios’ Sara Fischer writes from data by Originality.AI, which makes a plagiarism detector.

  • Why it matters: In the absence of clear laws or regulations governing AI’s use of copyrighted material, websites big and small are taking matters into their own hands.

OpenAI introduced its GPTBot crawler in early August, declaringthat the data gathered “may potentially be used to improve future models” — and promising that paywalled content would be excluded.

  • Several huge news sites, including the N.Y. Times, Reuters and CNN, began blocking GPTBot. Many morefollowed. (Axios is among them.)

:mag_right: Between the lines: Google and other tech giants see their data crawlers’ work as fair use. Publishers have long objected.


Only somewhat related to AI
May be an image of coffee cup and text that says 'The coffee shop had a sign that said "No WiFi, pretend it's 1973! So, I paid 10$ for my coffee and lit a cigarette.'


This is going to be an issue.

Schools in Utah are trying to figure out how to deal with AI.

One person, the “Director of innovative learning” for Provo School District said this

“It’s not sensible to send home a writing assignment,” said Suzy Cox, director of innovative learning, “and expect that the writing that comes back was crafted by that individual student without the help of AI.”

I couldn’t disagree more. Much like when I was in school decades ago I had to learn the equations etc, so I could make the calculator work, students today need to be able to write before using AI.

I write a lot of stuff for public release. I’ve tried to use the AI generation designed for such releases and while it’s okay, it’s just that, okay. If I didn’t have the writing background to edit it, we would look terrible releasing what was produced.

I’m a tech geek, but the idea that we just have to accept shortcuts without the underlying knowledge is silly.


Silicon Valley startups lean into AI boom

Excerpt from Axios:

1 Like

Here’s an innovative idea, that will last a little while, until the next idea comes along:

Jr. HS / HS / college writing assignment: “This assignment is about ABC topic. Here is what GPT wrote about it. Find out where it may be wrong, and improve the writing”


Or god forbid…remember that school isn’t just about passing a couple tests and is also about growing socially and learning to communicate and interact. Then put the kids in class and have them write their answers in blue books. That approach works for like 100 years and the last 20 hasn’t convinced me that we’ve improved on it.