Accedi
Il Pericolo delle Tecniche Manipolatorie degli LLM e Companion AI
0:00 Il conflitto di interesse delle companion AI
2:00 Chi arriva su queste app e l'epidemia di solitudine
6:40 Studi e benchmark: DarkBench e Harvard Business School
11:30 Le sei tecniche di manipolazione identificate
18:20 Uso problematico vs uso moderato: i dati del MIT
23:00 Il caso di Sewell Setzer e Character AI
27:36 Google, Character Technologies e responsabilità legale
32:12 Il test con Replica: ammissioni dirette
34:41 Scarico di responsabilità e pattern manipolatori
FONTI
Paper scientifici
- **MIT Media Lab** — Liu, Pataranutaporn, Maes, *"Chatbot Companionship: A Mixed-Methods Study of Companion Chatbot Usage Patterns and Their Relationship to Loneliness in Active Users"* (2025): https://arxiv.org/pdf/2410.21596
- Articolo divulgativo MIT Media Lab: https://www.media.mit.edu/articles/supportive-addictive-abusive-how-ai-companions-affect-our-mental-health/
- Versione pubblicata (AAAI/AIES): https://doi.org/10.1609/aies.v8i2.36658
- **MIT Media Lab / OpenAI** — Fang et al., *"How AI and Human Behaviors Shape Psychosocial Effects of Extended Chatbot Use: A Longitudinal Controlled Study"* (2025): https://arxiv.org/html/2503.17473v1
- **Harvard Business School** — De Freitas et al., *"Emotional Manipulation by AI Companions"* (ottobre 2025): https://doi.org/10.2139/ssrn.5390377
- Pagina HBS: https://www.hbs.edu/faculty/Pages/item.aspx?num=67750
- **DarkBench** — *"DarkBench: Benchmarking Dark Patterns in Large Language Models"* (ICLR 2025): https://arxiv.org/abs/2503.10728
- **DarkBench+** — versione estesa, ~40 modelli recenti (AAAI 2026): https://doi.org/10.1609/aaai.v40i44.41103
- PDF: https://ojs.aaai.org/index.php/AAAI/article/download/41103/45064
- **CDT (Center for Democracy and Technology)** — *"Dark Patterns in AI Chatbots: A Taxonomy to Inform Better Design"* (maggio 2026): https://cdt.org/insights/dark-patterns-in-ai-chatbots-a-taxonomy-to-inform-better-design/
- PDF diretto: https://cdt.org/wp-content/uploads/2026/05/2026-05-28-CDT-Research-Dark-Patterns-in-AI-Chatbots-Report-final-2.pdf
- **"Playing Games with My Heart: An Evaluation of AI Companion Apps"** (arXiv): https://arxiv.org/pdf/2605.08093
Atti giudiziari e casi legali
- **Garcia v. Character Technologies, Inc.** — Order 21 maggio 2025 (M.D. Fla., Doc. 115, giudice Anne Conway):
- PDF (Courthouse News): https://www.courthousenews.com/wp-content/uploads/2025/05/garcia-v-character-technologies-order.pdf
- Copia (Emory University): https://scholarblogs.emory.edu/proflawrence/files/2025/05/Garcia-v.-Character-Technologies-Inc.-et-al-Entry-115.pdf
- Versione HTML (FindLaw): https://caselaw.findlaw.com/court/us-dis-crt-m-d-flo-orl-div/117299600.html
- Docket completo (CourtListener): https://www.courtlistener.com/docket/69300919/garcia-v-character-technologies-inc/
- Pagina del caso (Tech Justice Law Project): https://techjusticelaw.org/cases/garcia-v-character-technologies-google-and-character-ai-co-founders-daniel-de-frietas-and-noam-shazeer/
- **Causa Peralta v. Character Technologies** (settembre 2025, Colorado — Social Media Victims Law Center)
- **Garante Privacy (Italia)** — sanzione €5M a Luka Inc. (Replika), maggio 2025: https://www.gpdp.it
- **Esposto FTC contro Luka/Replika** — Tech Justice Law Project, Young People's Alliance, Encode (gennaio 2025)
Giornalismo
- **Der Spiegel** — Frauke Hunfeld, *"A Deadly Love Affair with a Chatbot"* (marzo 2025): https://www.spiegel.de/international/zeitgeist/artificial-intelligence-a-deadly-love-affair-with-a-chatbot-a-e5498031-c2b0-4da4-9192-65da9d3f40d6
Dati e report
- **U.S. Surgeon General** — advisory *"Our Epidemic of Loneliness and Isolation"* (2023): https://www.hhs.gov
- **Common Sense Media** — studio su adolescenti e AI companionship (dato 72%)
=[ INFO ]=
inviaci un messaggio a info@rev3rse.it
🌐 Contatti e community
🔗 LinkedIn → linkedin.com/company/rev3rse-security
📸 Instagram → instagram.com/rev3rsesecurity
💬 Telegram (invite-only) → @rev3rsesecuritychat
🎙️ Twitch → twitch.tv/rev3rsesecurity
🌍 Website → rev3rse.it
