Vincent Lalanne

Enseignant chercheur - Universite de Pau et des Pays de l'Adour
Informatique - Cybersecurité

[Nav Toor @heynavtoor]

You have noticed it. ChatGPT feels dumber than it used to. Your prompts that worked six months ago produce worse results now. The writing sounds flatter. The ideas sound safer. The internet itself feels like it is shrinking. Every article reads the same. Every email sounds the same. Every answer sounds like it was written by the same voice.

You thought it was you. It is not you.

Researchers at Oxford and Cambridge published a paper in Nature proving what is happening. They call it Model Collapse.

Here is the mechanism in one sentence. AI trained on AI-generated data gets dumber every generation until it forgets what real human data looked like.

The internet is filling with AI-generated content. Blog posts. Articles. Reviews. Comments. Social media. AI companies scrape the internet to train the next generation of models. Which means the next generation of AI is being trained on the output of the current generation.

Each cycle loses information. Not randomly. It loses the rarest, most unusual, most creative parts first. The researchers call these the "tails of the distribution." The weird ideas. The unexpected perspectives. The things that made the internet feel human. Those disappear first.

What remains is the average. The safe. The expected. The bland.

Then the next generation trains on that. And loses more. And the next generation trains on that. And loses more. The researchers proved this is not a slow decline. Major degradation happens within just a few iterations. Even when some of the original human data is preserved.

They tested it on large language models. On image generators. On statistical models. The pattern was the same every time. The output converges toward a narrow, flattened version of reality that looks nothing like the original data.

The lead researcher put it plainly. "Large language models are like fire. A useful tool. But one that pollutes the environment."

The pollution is invisible. You cannot see which sentence on the internet was written by a human and which was written by AI. Neither can the AI that is about to train on it. And once the tails are gone, they do not come back. The damage is irreversible.

This is not a prediction anymore. It is a diagnosis.

The internet you grew up on was built by humans writing things no algorithm would have written. Strange, personal, imperfect, alive. That internet is being diluted. One generation of AI at a time. And the models trained on what remains are learning a smaller and smaller version of the world.

Model Collapse is not a technical problem. It is a cultural one. The thing that made the internet worth reading is the thing that disappears first.

[Elon Musk @elonmusk]

SpaceX is actively hiring world-class engineers/physicists for SpaceXAI, even if you have zero prior experience in AI. Smart humans figure it out fast.

Please send an email with ~3 bullet points demonstrating evidence of exceptional ability to [email protected].

[Vincent Lalanne @lalannev]

@Alokkumarzz @HeyAmit_ Agent

[Alex Veremeyenko @alex_verem]

🚨 BREAKING: Researchers at UW Allen School and Stanford just ran the largest study ever on AI creative diversity.

70+ AI models were given the same open-ended questions. They all gave the same answers.

They asked over 70 different LLMs the exact same open-ended questions.

"Write a poem about time." "Suggest startup ideas." "Give me life advice."

Questions where there is no single right answer. Questions where 10 different humans would give you 10 completely different responses.

Instead, 70+ models from every major AI company converged on almost identical outputs. Different architectures. Different training data. Different companies. Same ideas. Same structures. Same metaphors.

They named this phenomenon the "Artificial Hivemind." And the paper won the NeurIPS 2025 Best Paper Award, which is the highest recognition in AI research, handed to a small number of papers out of thousands of submissions.

This is not a blog post or a hot take. This is award-winning, peer-reviewed science confirming something massive is broken.

The team built a dataset called Infinity-Chat with 26,000 real-world, open-ended queries and over 31,000 human preference annotations. Not toy benchmarks. Not math problems.

Real questions people actually ask chatbots every single day, organized into 6 categories and 17 subcategories covering creative writing, brainstorming, speculative scenarios, and more.

They ran all of these across 70+ open and closed-source models and measured the diversity of what came back. Two findings hit hard.

First, intra-model repetition. Ask the same model the same open-ended question five times and you get almost the same answer five times.

The "creativity" you think you're getting is the same output wearing a slightly different outfit. You ask ChatGPT, Claude, or Gemini to write you a poem about time and you keep getting the same river metaphor, the same hourglass imagery, the same reflection on mortality.

Over and over. The model isn't thinking. It's defaulting to whatever scored highest during alignment training.

Second, and this is the one that should really alarm you, inter-model homogeneity. Ask GPT, Claude, Gemini, DeepSeek, Qwen, Llama, and dozens of other models the same creative question, and they all converge on strikingly similar responses.

These are models built by completely different companies with different architectures and different training pipelines.

They should be producing wildly different outputs. They're not. 70+ models all thinking inside the same invisible box, producing the same safe, consensus-approved content that blends together into one indistinguishable voice.

So why is this happening? The researchers point directly at RLHF and current alignment techniques. The process we use to make AI "helpful and harmless" is also making it generic and boring.

When every model gets trained to optimize for human preference scores, and those preference datasets converge on a narrow definition of what "good" looks like, every model learns to produce the same safe, agreeable output. The weird answers get penalized.

The original takes get shaved off. The genuinely creative responses get killed during training because they didn't match what the average annotator rated highly. And it gets even worse.

The study found that reward models and LLM-as-judge systems are actively miscalibrated when evaluating diverse outputs. When a response is genuinely different from the mainstream but still high quality, these automated systems rate it LOWER. The very tools we built to evaluate AI quality are punishing originality and rewarding sameness.

Think about what this means if you use AI for brainstorming, content creation, business strategy, or literally any task where you need multiple perspectives. You're getting the illusion of diversity, not the real thing.

You ask for 10 startup ideas and you get 10 variations of the same 3 ideas the model learned were "safe" during training. You ask for creative writing and you get the same therapeutic, perfectly balanced, utterly forgettable tone that every other model gives.

The researchers flagged direct implications for AI in science, medicine, education, and decision support, all domains where diverse reasoning is not a nice-to-have but a requirement.

Correlated errors across models means if one AI gets something wrong, they might ALL get it wrong the same way. Shared blind spots at massive scale.

And the long-term risk is even scarier. If billions of people interact with AI systems that all think identically, and those interactions shape how people write, brainstorm, and make decisions every day, we risk a slow, invisible homogenization of human thought itself. Not because AI replaced creativity.

Because it quietly narrowed what we were exposed to until we all started thinking the same way too.

Here's what you can actually do about it right now:
→ Stop accepting first-draft AI output as creative or diverse. If you need 10 ideas, generate 30 and throw away the obvious ones
→ Use temperature and sampling parameters aggressively to push models out of their comfort zone
→ Cross-reference multiple models AND multiple prompting strategies, because same model with different prompts often beats different models with the same prompt
→ Add constraints that force novelty like "give me ideas that a traditional investor would hate" instead of "give me creative ideas"
→ Use structured prompting techniques like Verbalized Sampling to force the model to explore low-probability outputs instead of defaulting to consensus
→ Layer your own taste and judgment on top of everything AI gives you. The model gets you raw material. Your weirdness and experience make it original

This paper puts hard data behind something a lot of us have been feeling for a while. AI is getting more capable and more homogeneous at the same time.

The models are smarter, but they're all smart in the exact same way. The Artificial Hivemind is not a bug in one model. It's a systemic feature of how the entire industry builds, aligns, and evaluates language models right now.

The fix requires rethinking alignment itself, moving toward what the researchers call "pluralistic alignment" where models get rewarded for producing diverse distributions of valid answers instead of collapsing to a single consensus mode.

Until that happens, your best defense is awareness and better prompting.

[Vincent Lalanne @lalannev]

@bearstech Y a 30 ans le PC familial avait deux disquettes pour le boot : une sur win95 (jeux des enfants) et l'autre sur slackware. Un jour mon fils de 3 ans qui se préparait à son petit jeu d'énigmes hurle dans l'appartement " Papa t'as booté sur Linux ! "

[Ilan Gabet @Ilangabet]

THREAD Marine Le Pen condamnée à 4 ans de prison et 5 ans d’inéligibilité avec exécution provisoire.

Le FN va lancer une guerre de communication pour manipuler l’opinion.

Je vous fournis donc un décryptage de l’ensemble de leurs mensonges pour que nous jouions à armes égales ⬇️

[Vincent Lalanne @lalannev]

@SectionPaloise @FestivaldeSiros
Ce soir au festival nous chantons la victoire de la section !

[Vincent Lalanne @lalannev]

Félicitations Angèle !
@universite_uppa

[Ministère Enseignement supérieur/Recherche/Espace @sup_recherche]

#EquipeDesFrançais #Paris2024

👏🥈 Félicitations à Angèle Hug, étudiante sportive de haut niveau en M2 STAPS à @universite_uppa qui remporte une médaille d’argent en canoë-kayak ! 🥈👏

Cc @Paris2024 @Sports_gouv @gouvernementFR @sretailleau

[Anna Colin Lebedev @colinlebedev]

Comment s’installe un pouvoir liberticide?

Dix leçons tirées de l’observation des évolutions politiques de la Russie.
1/20, 🧶

[Guillaume Limousin @Sonic_urticant]

🔥📣[THREAD]
1/ En exclusivité (même les journalistes attendaient cela mais je le dévoile sur Twitter) : le récit détaillé de mon procès du 1er mars 2024 qu'avait lancé Didier Raoult contre moi (il a perdu).
Récit préalablement relu par mes avocats.
Et il y a du lourd...
RT SVP⏬

[Vincent Lalanne @lalannev]

@Korben Bon anniversaire !

[Vincent Lalanne @lalannev]

@NicolasDOM19857 https://t.co/DtzqdPoqWP

[Cédric C. (Prof) @teamaaz]

Le 26/02/24 @bluetouff interpelle @FranceTravail sur une vulnérabilité.

J'imagine qu'il ne reçoit aucune réponse.

Le 13/03 @FranceTravail annonce avoir subi une cyberattaque visant "potentiellement" 43 millions de personnes sur son site officiel.

C'est fatiguant...

[🏴‍☠️ MC Pablo ∴ #SyndiquezVous 🇵🇸 @p4bl0]

@claireplacial Dans le genre : Opération Postes, pour les sections CNU 25, 26, et 27 (maths et info).
https://t.co/4qcdPirz6Q

Sur la page il y a aussi les liens vers les initiatives similaires (wiki auditions) pour les sections : 17, 19, 20, 21 & 22, 23, 70, et 74. #ESR #campagne

[Vincent Lalanne @lalannev]

@LSetterahmane @GabrielAttal @gouvernementFR @auroreberge @CaVautrin Merci Simon•e 🙂

[guy birenbaum @guybirenbaum]

RT si tu fais partie de ces gamin(e)s du public qui doivent tout à l'Éducation nationale.

[Vincent Lalanne @lalannev]

Par contre je vous certifie que @FrPellegrini est réel !

[Tansu Yegen @TansuYegen]

Model Emily Pellegrini, amassing 137,000 followers in 4 months, is a digital creation. Formed by neural networks as the 'average man's dream girl,' she's a virtual Instagram hit, earning $10,000 and attracting billionaire suitors, highlighting the era of digital illusion. 👩‍🎤
#CES2024

[Le Libre Éducatif @LeLibreEdu]

[Emploi du temps libre et ergonomique]
"flop!EDT est un logiciel libre de génération automatique et de gestion coopérative des emplois du temps (scolaires ou planning de salarié⋅e⋅s). Il est distribué sous licence AGPLv3."
Développé à @iutblagnac
https://t.co/dfOZySOaUR

[association EPI @assoEPI]

Deux postes de PRAG s'ouvrent à Castres en septembre 2024. Clôture des candidatures le 23/11/2023 :
https://t.co/y7TLV7Nd69

https://t.co/HpRnnACrpR