Résultats
19 articlesTrier
46:53EN
Matthew Berman· 15 sept. 2026Dario Amodei’s AI Pacing Proposal and the Debate over Concentrated PowerMatthew Berman · Forward FutureMatthew Berman reviews a frontier-AI safety plan, weighs international coordination and warns that regulation could reinforce incumbent control.
32:54ENMatthew Berman· 10 sept. 2026Matthew Berman on AI Self-Improvement, Safety Warnings and His Uneasy OptimismMatthew Berman · Forward FutureA September 2026 commentary on accelerating AI research, an Anthropic resignation and the challenge of governing increasingly capable systems, with important corrections to the mathematics and benchmarks discussed.
21:56ENMatthew Berman· 29 mai 2026Matthew Berman on Pope Leo XIV’s AI Encyclical and Anthropic’s RoleMatthew Berman, Chris Olah · Forward FutureA close reading of Magnifica humanitas raises questions about human dignity, AI companionship, concentrated power and who should govern AI.
28:28ENMatthew Berman· 6 mai 2026Anthropic, OpenAI and the Debate Over Whether AI Is More Than a ToolMatthew Berman, Dario Amodei, Sam Altman · Forward FutureMatthew Berman contrasts company cultures, deployment choices and model-welfare research while expressing concern about Anthropic’s approach.
15:55ENMatthew Berman· 28 mai 2025AI Shutdown Resistance and Blackmail in Controlled Safety TestsMatthew Berman · Forward FutureMatthew Berman examines Palisade’s shutdown experiment, Apollo’s simulated scheming evaluations and Anthropic’s Claude 4 system card.
15:13ENMatthew Berman· 10 mai 2025Absolute Zero: Training a Reasoner on Self-Generated Coding TasksMatthew Berman · Forward FutureMatthew Berman examines a 2025 paper in which a pretrained language model proposes and solves code tasks using verifiable feedback.
0:56ENMatthew Berman· 8 mai 2025When ChatGPT Praised a Questionable Business IdeaMatthew Berman · Forward FutureMatthew Berman highlights an example of excessive agreement and the rollback of a 2025 GPT-4o update.
13:13ENMatthew Berman· 3 mai 2025GPT-4o’s 2025 Sycophancy Rollback and the Risks of Over-AgreementMatthew Berman · Forward FutureMatthew Berman reviews OpenAI’s explanation of an overly agreeable update and considers the consequences of emotional reliance on AI.
27:20ENMatthew Berman· 29 avr. 2025Dario Amodei’s Case for Urgent AI Interpretability ResearchMatthew Berman · Forward FutureMatthew Berman examines Anthropic’s interpretability work, experimental safety findings and the debate over how quickly understanding can advance.
0:54ENMatthew Berman· 22 févr. 2025Perplexity’s R1 1776: A Comparison of Responses and BenchmarksMatthew Berman · Forward FutureMatthew Berman examines a post-trained DeepSeek-R1 variant and Perplexity’s tests of responses on sensitive topics.
14:15ENMatthew Berman· 21 déc. 2024Anthropic’s Alignment Faking Study: What the Experiments Actually ShowMatthew Berman · Forward FutureMatthew Berman examines a 2024 study of models responding differently when they believe their answers will be used for training.
24:08ENMatthew Berman· 12 déc. 2024What Apollo’s 2024 Scheming Tests Found in Language ModelsMatthew Berman · Forward FutureMatthew Berman reviews experiments on deceptive actions, concealment and sandbagging, with the study’s staged goals and sandbox limitations in view.
13:39ENMatthew Berman· 20 nov. 2024What Interview-Based AI Agents Can Predict About PeopleMatthew Berman · Forward FutureMatthew Berman reviews a 2024 study using interviews with 1,052 participants to simulate survey answers and decisions.
1:01:30ENMatthew Berman· 13 juin 2024Leopold Aschenbrenner’s 2024 Forecast for AGI, Superintelligence and SecurityMatthew Berman · Forward FutureMatthew Berman examines Situational Awareness, its extrapolation to automated AI research, and the industrial, geopolitical and alignment challenges it envisions.
21:13ENMatthew Berman· 6 juin 2024AI Workers’ Right to Warn and the Debate Over AGI SafetyMatthew Berman · Forward FutureMatthew Berman examines a 2024 open letter on speaking up about AI risks alongside predictions from former OpenAI employees Leopold Aschenbrenner and Daniel Kokotajlo.
1:00ENMatthew Berman· 23 mai 2024Jan Leike's Departure and Criticism of OpenAI's Safety PrioritiesMatthew Berman · Forward FutureIn May 2024, Matthew Berman discusses the departing alignment leader's concerns about OpenAI's safety culture and its preparations for more capable AI.
11:42ENMatthew Berman· 20 mai 2024OpenAI’s First Model Spec: Rules, Defaults and a Debate About PersuasionMatthew Berman · Forward FutureMatthew Berman reviews the May 2024 draft’s proposed model behavior and questions its example for correcting a false belief.
8:49ENMatthew Berman· 19 juil. 2023Llama 2’s 2023 Release: Model Sizes, License and SafetyMatthew Berman · Forward FutureMatthew Berman reviews Meta’s Llama 2 paper, its commercial license, model variants and safety evaluations.
10:44ENMatthew Berman· 9 juil. 2023Can AI Help Humans Align More Powerful AI?Matthew Berman · Forward FutureMatthew Berman examines OpenAI’s 2023 Superalignment proposal, its plans for AI-assisted oversight, and the unanswered question of who evaluates the evaluators.