The DU Lounge
Related: Culture Forums, Support ForumsA very sweet Christmas commercial - with real actors, no AI animation or AI voiceover
This is presently airing on French TV for Intermarche, a super market chain, and tells the story of how a wolf changes his diet.
"Manger mieux" - "eat better", being a pitch for less meat and more vegetables.
Irish_Dem
(79,809 posts)From AI:
"Intermarché Conte de Noël" translates to "Intermarché Christmas Tale" or "Intermarché Christmas Story,"
referring to the French supermarket's heartwarming animated 2025 holiday commercial about a wolf learning to cook vegetarian food to make friends, promoting themes of acceptance and kindness through food.
Here's a breakdown of the title:
Intermarché: The name of the French supermarket chain.
Conte: French for "tale" or "story" (like a fairy tale).
de Noël: French for "of Christmas" or "Christmas".
The commercial tells an emotional Christmas Story about a wolf's journey of change and friendship,
using Intermarché products as the key to his transformation.
GoneOffShore
(17,979 posts)Irish_Dem
(79,809 posts)highplainsdem
(60,002 posts)Irish_Dem
(79,809 posts)highplainsdem
(60,002 posts)you'd watch the video and do your own summary, not copy whatever some hallucination-prone software offered as a summary.
I've seen a lot of threads online raving about how great this commercial is, especially compared to recent AI slop like the McDonalds commercial that aired only in the Netherlands but got worldwide backlash and was quickly yanked: https://futurism.com/artificial-intelligence/mcdonalds-ai-generated-commercial . No one discussing the French commercial in the OP seemed to think a commercial that's a strong argument against AI needed an AI summary.
AI summaries are also slop (and very damaging to the internet - https://www.theguardian.com/technology/2025/jul/24/ai-summaries-causing-devastating-drop-in-online-news-audiences-study-finds ), and posting one in a discussion of animation that's good because it isn't AI looks like a vote for AI.
Irish_Dem
(79,809 posts)1. I am fluent in basic French language. I took French classes starting in 5th to 12th grade
and then two years of advanced French in College. I also have French speakers in my family.
2. I watched the video myself.
3. I have a PhD and am trained as a scientist. I ALWAYS review primary and secondary source data
for accuracy. I am horrified to be accused of publishing SLOP. What an insult to me, my integrity,
and to my PhD. Which I worked damn hard for.
4. I made a post helping others enjoy the video and get attacked for it.
5. The anti AI people are just as bad as the AI lovers.
highplainsdem
(60,002 posts)internet and also often inaccurate - and if you aren't already expert in the subject of the AI summary, it's unlikely the AI errors will be caught. AI companies warn that their tools can always give wrong results and those results should be verified. But since people are accepting or requesting AI summaries in the first place to save time, that motivation makes it less likely they'll ever check the results against a more reliable source. I've seen teachers post about students who got incorrect information from ChatGPT and refused to believe it was wrong even when shown encyclopedias with the correct information.
So I believe using AI summaries as a source is very harmful. And then there's the damage to the internet those AI summaries are doing.
Btw, since you're trained as a scientist, you're probably aware that sites publishing scientific papers are now offering AI summaries of papers, ABOVE the abstracts, for people who can't be bothered to even read the short abstract. And according to posts I've seen from scientists, those AI summaries often get their LLM-generated info completely wrong. But no one will know that unless they ignore the summary and actually read beyond it.
So I HOPE you don't approve of AI summaries on those sites. And maybe that will help you understand why I considef them harmful, period.
Even though you felt the information would help people enjoy the video, you're not likely to find something spat out by a genAI tool enjoyed in any thread about appreciating human-created work that didn't use AI (and the commercial NOT using AI was a point the OP specifically made). Just as it wouldn't have fit in if you'd decided to generate AI video of an audience wildly applauding that human-made commercial. It misses the point.
The anti-AI people are trying to stop greedy, unethical AI companies from stealing the world's intellectual property, sabotaging education and the economy and politics, enshittifying the information ecosystem, and harming the environment. Those are worthwhile goals. Liberal goals. They don't become any less liberal when AI companies and proponents are doing the damage.
And even though AI results can at times be accurate, it's all slop because using genAI is a technological crap shoot with no real intelligence behind it. As some experts say, LLMs are always hallucinating, but some of those hallucinations are more accurate than others. Some other experts call these AI tools bullshit machines.
Humans can do better, and forums deserve better, than genAI. Do you realize how fast this board would fill up with AI-generated replies if it was considered okay here? It could turn into what in effect would be a bot message board.
highplainsdem
(60,002 posts)If you said you felt insulted for that reason, you misread my comments.
But since you do know French, it would have been even more interesting to hear your own take on the commercial, perhaps translating more than just the title. Rather than just seeing an AI summary you copied.
I wasn't insulting your degree or education. I was pointing out what's wrong with AI summaries.
And the term "slop" is often used for AI summaries. Your having a PhD and deciding an AI summary is OK to post does not change the fact that it's AI slop. That it happens to be accurate doesn't change the fact that it's AI slop. It's the source that matters. The AI tool, which does not think and isn't intelligent, can offer very different answers to the same prompt - simultaneously to a single user, if whoever controls the tool allows multiple responses.