I Asked AI How to stop Trump. And Here's What It Said... (From ABC news)

[marble falls]

?si=3D4EUwscfBmuqV9N

[CousinIT]

1. Skip 7:40 into the video to get beyond all the introductory BS. SO annoying. Most of the video

is "blah blah blah..."

The actual meat of it is after 7:40 and there's plenty of cruft and crust after that too.

If people are going to do videos like this, they need to get to it and avoid the blabbering and cruft. Nobody's going to watch all of that.

[highplainsdem]

4. Skip it all. It's just AI slop from someone who started out with AI videos a year ago and then

discovered he could hook more gullible viewers with videos on politics than with videos on "What if dragons really existed?"

He'd still be posting about dragons if that got more views.

[Polybius]

17. I think if dragons existed, we might have to destroy them

I'd imagine that they would be a huge danger to planes.

[highplainsdem]

18. True. But I wouldn't be surprised if there are people out there wondering if selective breeding of Draco lizards

might eventually produce some more than 8 feet long, rather than 8 inches long.

https://en.wikipedia.org/wiki/Draco_(lizard)

[LAS14]

2. How about a summary? nt

[highplainsdem]

5. AI slop isn't worth summarizing. The YouTuber who did this started a year ago doing AI videos on

different subjects, then discovered more people would fall for his AI slop if it was about politics instead of dragons or superheroes or time travel.

[Trust_Reality]

10. Except for some promotional stuff this is a summary.

Isn't that kind of what AI does?

[LAS14]

19. I don't want to watch a video for 7 minutes. nt

[highplainsdem]

3. This is AI slop. And it is NOT from ABC News, so your thread title is completely misleading. This video

was done by one more YouTube and AI exploiter who got that YouTube channel only a year ago and was posting AI garbage videos like "What if dragons really existed?" and "What if internet shut down forever?"

Then he discovered he could get a lot more views for his AI slop if he used politics as a subject.

YouTube is filling up with all kinds of AI crap. I hope DU won't, because the wankers uploading these videos can churn out lots of videos every day to waste people's time, maybe make some money, and divert people's attention from human reporting and expert opinion on Trump.

[SharonClark]

6. I clicked on the link and watched the whole thing, something I never do when no additional

information is included by the poster. I guess I'm being generous with my time this morning.



The answers were common sense statements of our current situation and how to fight trump. Withou the fluff and filler, it would make a good introduction to those newly interested in the fight for Democracy.

A transcript would be helpful.

[highplainsdem]

7. There's plenty of good advice out there from real experts. This video is just from an exploiter who started

out using AI to post on other subjects (dragons, superheroes, time travel), then found a larger audience with politics. The person doing these videos may not care at all about politics. His videos are pure clickbait and he does not deserve an audience for them.

Watching his crap takes time from reading or watching the journalists and experts who really care about politics and really know what they're talking about.

[Trust_Reality]

9. From what I can tell, that good advice from real experts influenced this presentation quite a bit.

I think it is VERY INERESTING though not earth shakingly new.

[highplainsdem]

14. The video itself is very harmful. Generative AI like ChatGPT is unethically and illegally trained on

stolen intellectual property.

The video is fundamentally dishonest in saying ChatGPT has a plan.

The YouTuber who produced that video is a cynical exploiter of YouTube, AI and people's gullibility, since he switched to politics as a subject when dragons, superheroes and time travel didn't get as many views for him.

It's a waste of people's time that could otherwise be spent on real advice from real people whose news stories, editorials, podcasts and YouTube videos deserve attention, when this AI slop doesn't.

And to top off the problems, marble falls says in the thread title that it's from ABC News, and it isn't.

[et tu]

8. interesting

even ai can understand what a dictatorship is in our country.
thank you for posting~

[highplainsdem]

13. AI doesn't "understand" anything. Generative AI like ChatGPT can give very different answers to the

exact same prompt at different times - or even at the same time, if set up to offer multiple answers, as ChatGPT could be but hasn't been, probably because these chatbots look much less authoritative if users see them offering different answers in the same persuasive style. If you've ever used an AI image generator, or seen articles about them, you've probably seen that they offer multiple options from the same prompt, often wildly different. The chatbot has no awareness of what it's offering. The same is true with text output. The AI peddlers just don't want to make people aware of that, usually.

See this earlier post - #13 there - with a link to an article on chatbots offering different answers to the same question.

https://www.democraticunderground.com/100220334521#post13

There are real authorities on Trump and dictatorships out there - human authorities, reliable sources. It's a waste of time to use AI. And a waste of time and attention to reward a YouTube clickbait producer who did videos on dragons and superheroes before deciding to use politics as bait. That YouTuber might have no real interest in politics. He's just using AI to try out different bait.

[WhiskeyGrinder]

12. Fuck this slop.

[highplainsdem]

15. +1,000,000,000,000

[andym]

16. Just states the obvious-- there is a transcript at youtube: 8:20-13:59 is the relevant part. Including an AI summary

For fun I asked Google's Gemini to summarise the transcript from 8:20-13:59 because the transcript is difficult to read:

Key points of the proposed plan by ChatGPT include:

State-Level Resistance: Blue and swing states should use their powers to counter federal actions. This includes state attorneys general suing the federal government, governors refusing to cooperate with federal programs or blocking their implementation, and state legislators passing counter laws and denying resources.
Media Warfare: The focus should be on reaching undecided individuals through trusted local news and aggressive, emotionally resonant digital content (TikTok, memes, YouTube, podcasts) to combat disinformation and highlight corruption.
Civil Disobedience: If the system is unresponsive, the plan calls for peaceful but forceful actions like mass strikes by public and private workers, boycotts of Trump-aligned corporations, and disruption campaigns such as legally occupying government buildings and overwhelming the courts.
Contingency Planning (Survival Plan) In a scenario (nearing autocracy), the plan suggests establishing encrypted communications, having legal aid teams ready, and creating safe havens for targeted groups like journalists, transgender children, and whistleblowers.
Grassroots Political Engagement: Building a counter-movement from the local level up is emphasized. This involves running for local offices (school boards, city councils), identifying and supporting new political voices, and making politics personal by addressing people's immediate needs like jobs and childcare.
International Pressure: Leveraging global opinion by encouraging allies to withdraw support, seeking UN condemnation, disrupting trade deals, prompting foreign divestment, and using international courts to expose abuses.
Impeachment (Considered a Long Shot): While technically an option, impeachment is viewed as unlikely to succeed and problematic even if it did. The transcript mentions that with "JD Vance as vice president," removing Trump would just hand power to someone potentially "just as hardline, maybe even more ruthless." It's only deemed viable with overwhelming political momentum and a clear follow-up strategy.

Regarding ChatGPT's supposed belief in its own plan:

Conditional Efficacy: The AI purportedly states the plan could work if people were highly organized, committed, and acted with urgency.
Pessimism about Public Engagement: However, it expresses doubt that most people would undertake such efforts, predicting many would limit their actions to complaining, online posts, or voting, while others would disengage.
Opponent's Dedication: It contrasts this with the perceived full commitment, funding, and long-term focus of "the people pushing the agenda."
Low Likelihood of Full Success: The AI doesn't believe enough people will act decisively to "stop it cold" unless a major event galvanizes them.
Hope in Partial Success: A more likely scenario, it suggests, is that enough people might act by 2026 to achieve limited gains, such as flipping key races or slowing down the agenda, which is framed as "something... a whole lot better than nothing."
Overall Stance: The AI doesn't "count on the plan" for complete success but holds onto a "long shot" hope that enough people will care in time, expressing a wish to be proven wrong in its pessimism.

----
Keep in mind that the video creator is anthropomorphising the AI. The AI doesn't believe anything-- it's not how it works.