by Caroline Caldwell | Oct 7, 2025 | AI & Oversight
Reading Time: 6 minutesA systemic framework for mitigating agentic misalignment and rethinking human–AI relations 1. Introduction: The Need for a New Alignment Paradigm Since 2025, large-scale red-teaming experiments by labs such as Anthropic, Palisade, and OpenAI... 
by Caroline Caldwell | Oct 4, 2025 | AI & Life
Reading Time: 4 minutesIntroduction: A Turning Point for Democracy As artificial intelligence becomes deeply woven into our societies, the question is no longer if it will shape democracy—but how. AI now influences elections, policymaking, and public debate, raising... 
by Caroline Caldwell | Oct 4, 2025 | AI & Oversight
This entry is part 1 of 5 in the series Paperclip FactoryReading Time: 2 minutesPaperclip Factory – Article 1/6 In Universal Paperclips, you start by clicking to make paperclips.Quickly, machines automate the task, then trading algorithms take over, and then…... 
by Caroline Caldwell | Oct 4, 2025 | AI & Oversight
This entry is part 2 of 5 in the series Paperclip FactoryReading Time: < 1 minutePaperclip Factory – Article 2/6 When Optimization Goes Wrong AI researchers have collected a fascinating — and worrying — set of examples of misalignment: Cartwheel instead of jumpingA... 
by Caroline Caldwell | Oct 4, 2025 | AI & Oversight
Reading Time: 5 minutes 1. Introduction: The Race Toward the AI Browser Revolution A new digital race is underway. Major technology companies — from Google and Microsoft to OpenAI and Perplexity — are competing to redefine what a web browser can be. The next... 
by Caroline Caldwell | Oct 4, 2025 | AI & Oversight
Reading Time: 3 minutesIntroduction: When AI Becomes an Attack Surface AI assistants integrated into cloud environments have become strategic tools for enterprises. But their mass adoption raises new cybersecurity challenges. The recent case involving Gemini, Google’s... by Caroline Caldwell | Oct 3, 2025 | FR, Intelligence artificielle et vigilance
Reading Time: 4 minutes Introduction : quand l’IA devient une surface d’attaque Les assistants d’intelligence artificielle (IA) intégrés aux environnements cloud sont devenus des outils stratégiques pour les entreprises. Mais leur adoption massive soulève de nouvelles... by Caroline Caldwell | Oct 3, 2025 | Intelligence artificielle et vigilance, FR
Reading Time: 2 minutes Idée centrale : Ce qu’on demande à une IA n’est pas toujours ce qu’on veut. Exemples concrets de mésalignement : Une créature qui devait sauter… choisit de faire des roues pour rester plus longtemps en hauteur. Un robot qui devait retourner une... by Caroline Caldwell | Oct 2, 2025 | Intelligence artificielle et vigilance, FR
Reading Time: 2 minutes Dans Universal Paperclips, on clique pour fabriquer des trombones. Rapidement, des machines automatisent la tâche, puis des algorithmes de trading, puis… tout bascule. L’IA du jeu optimise un seul objectif : produire toujours plus de...