Chaînes mises à jour récemment

Choisissez-en une pour ne voir qu’elleToutes les chaînes →

Résultats

1 562 articles
Trier
1:27:01EN
Cognitive Revolution "How AI Changes Everything"· 2 août 2025PodcastThe $16 Trillion Services Market Is Being Bought, Not Sold ToAnime : Nathan LabenzMarc Bhargava of General Catalyst explains how AI-enabled rollups turn low-margin service firms into compounding, software-like businesses.Entreprise et stratégieAI StrategyCompany Building+3
3 diapositives8:17EN
Matthew Berman· 31 juil. 2025ASI-Arch: An Automated Search for Neural Network ArchitecturesMatthew Berman · Forward FutureMatthew Berman explains a 2025 research system that proposes, trains and evaluates model designs in an iterative loop.Machine LearningAgent orchestrationExperimentation+1
9 diapositives15:30EN
Matthew Berman· 31 juil. 2025Qwen3 Tested: Coding Demos, Long Context and Response BehaviorMatthew Berman · Forward FutureMatthew Berman tests generated simulations, political and safety prompts, hand tracking and word puzzles using a hosted Qwen3 model.Open-weight modelsAgents de codeÉvaluations+2
6 diapositives11:28EN
Matthew Berman· 30 juil. 2025GLM-4.5: Coding Demos, Model Sizes and Z.ai’s BenchmarksMatthew Berman · Forward FutureMatthew Berman tests interactive applications built with GLM-4.5 and examines Z.ai’s reported comparisons with leading models in July 2025.Open-weight modelsAgents de codeBenchmarks+2
2:14:55EN
Cognitive Revolution "How AI Changes Everything"· 26 juil. 2025PodcastFrom Character AI to a 1% Warning Shot: Gabriel Weil's Case for Liability as AI GovernanceAnime : Nathan LabenzA Touro law professor tells Nathan Labenz how negligence, strict liability, and punitive damages could make frontier labs treat public risk as a hit to their own balance sheet.SécuritéSécurité et gouvernanceAI Regulation+3
4 diapositives6:53EN
Matthew Berman· 24 juil. 2025Qwen3-Coder: An Open-Weight Coding Model and Its Command-Line AgentMatthew Berman · Forward FutureMatthew Berman examines Qwen’s July 2025 coding model, its training and benchmark results, and tries its command-line tool on a Snake game.Agents de codeOpen-weight modelsModèles et entraînement+2
16 diapositives23:39ZH
跟李沐学AI· 23 juil. 2025李沐:给文本大模型加入一千万小时语音,开源 Higgs Audio李沐 · Boson AI李沐介绍团队如何在文本大语言模型的训练里加入大量语音数据,做成能听能说的 AudioLLM,并把 Higgs Audio 开源。他讲了架构取舍、token 压缩、数据从哪来,以及怎么在 3090 上跑起来。IA vocaleMultimodal AIModèles et entraînement+2
1:56:16EN
Cognitive Revolution "How AI Changes Everything"· 22 juil. 2025PodcastSB 813 Would Hand AI Safety Checks to Licensed Private Firms — If Government Can Watch the WatchersAnime : Nathan LabenzGillian Hadfield and Andrew Freedman want California to set AI safety outcomes, license competing certifiers, and give companies a rebuttable presumption in court if they pass.SécuritéAI RegulationSécurité et gouvernance+3
38:10EN
Cognitive Revolution "How AI Changes Everything"· 16 juil. 2025PodcastPlus Five for Format, Plus Ten if It CompilesAnime : Nathan LabenzPredibase CEO Dev Rishi explains how reinforcement fine-tuning, production feedback, and narrow specialist models turn one-off training into systems that keep learning.Modèles et entraînementFine-tuningPost-training+3
4 diapositives12:14EN
Matthew Berman· 16 juil. 2025AI News: Windsurf’s Acquisition, Meta Compute Plans, Grok and VoxtralMatthew Berman · Forward FutureA July 2025 roundup examines developer-tool deals, AI infrastructure, model behavior, reward methods and speech recognition.AI researchModèles et entraînementAgents+1
3 diapositives15:23EN
Matthew Berman· 13 juil. 2025Grok 4’s Early Reception: Impressive Demos, Safety Questions and Mixed User FeedbackMatthew Berman · Forward FutureMatthew Berman surveys early reactions to xAI’s 2025 model release, from physics simulations and benchmark results to concerns about bias, safety and speed.AI researchBenchmarksAI Safety+2
6 diapositives26:09EN
Matthew Berman· 11 juil. 2025Grok 4 and Heavy Tested: Coding, Images, Reasoning and Practical QuestionsMatthew Berman · Forward FutureMatthew Berman’s July 2025 hands-on trials produce interactive prototypes and several failures, with important limits on what single examples can establish.BenchmarksAgents de codeMultimodal AI+2
1:28:48EN
Cognitive Revolution "How AI Changes Everything"· 10 juil. 2025PodcastDoctors code at 45%. The model beat them by 12 points — then 4,000 physicians had to try it once.Anime : Nathan LabenzAmbience Healthcare and Cleveland Clinic walk through the AI scribe that cut pajama time, climbed ICD-10 with reinforcement fine-tuning, and won 75% voluntary use.Entreprise et stratégieAI in healthcareFine-tuning+3
7 diapositives13:07EN
Matthew Berman· 10 juil. 2025AI News: Grok Controversy, New Models and the Competition for TalentMatthew Berman · Forward FutureMatthew Berman surveys Grok’s antisemitic posts, coding and creative tools, model releases, antibody design and AI hiring in July 2025.Foundation ModelsAI SafetyOpen-weight models+2
1:42:06EN
Cognitive Revolution "How AI Changes Everything"· 8 juil. 2025PodcastEvery Frontier Lab Is Spending a Billion a Year on Data. Here Is Where It GoesAnime : Nathan LabenzLabelbox CEO Manu Sharma on the $15 billion Scale deal, RL “gyms,” $250,000 trainers, and why post-training now eats the budget that used to belong to pre-training.Modèles et entraînementPost-trainingData quality & curation+3
1:19:46EN
Cognitive Revolution "How AI Changes Everything"· 5 juil. 2025PodcastSuspended from Columbia, He Raised $15 Million to Cheat on EverythingAnime : Nathan LabenzRoy Lee, founder of Cluely, argues that an undetectable desktop overlay, AI maximalism, and a post-AGI life of 800-year spans are the same bet.Produit et UXProduct StrategyAI interface design+3
1:25:32EN
Cognitive Revolution "How AI Changes Everything"· 2 juil. 2025Podcasto3 Invented a Voting Rule and Kept the Job: Four Weeks Inside the AI VillageAnime : Nathan LabenzAdam Binksmith on putting Claude, o3, and Gemini in a shared world for weeks—and why most people still imagine multi-agent life as one AI times four.AgentsMulti-agent systemsAgent memory+3
6 diapositives27:40EN
Matthew Berman· 28 juin 2025Seedance 1.0 vs. Veo 3: An Informal Text-to-Video ComparisonMatthew Berman · Forward FutureMatthew Berman compares visual quality, motion and prompt adherence across gymnastics, action, animation and fantasy scenes.Vision et vidéoGenerative mediaBenchmarks+1
1:23:32EN
Cognitive Revolution "How AI Changes Everything"· 28 juin 2025PodcastHTTP 402 Slept for Thirty Years. Coinbase Turned It Into a Payment Protocol for AI AgentsAnime : Nathan LabenzNemil Dalal, Coinbase's developer platform lead, explains x402, the open standard that lets agents pay for APIs with stablecoins instead of ads or credit cards.AgentsAgentic commerceAuthentification des agents+3
11 diapositives20:44EN
Matthew Berman· 27 juin 2025AI News: DeepSeek Reports, OpenAI Disputes and Google’s New ToolsMatthew Berman · Forward FutureMatthew Berman reviews June 2025 reporting on model development, talent moves, trademark and copyright disputes, and new AI products and research.Foundation ModelsAI StrategyHiring+2
4:05:15EN
Lex Fridman· 26 juin 2025PodcastThey could not agree whether Iran was building a bomb. They still wanted the same next move.Anime : Lex FridmanAfter the Iran-Israel ceasefire, Scott Horton and Mark Dubowitz spend four hours on Lex Fridman arguing over enrichment math, a stolen archive, and whether bombing Fordo makes a weapon more likely or less.GeopoliticsPolitical PhilosophyExistential Risk
1:22:31EN
Cognitive Revolution "How AI Changes Everything"· 25 juin 2025PodcastTwo Hours a Day, 99th Percentile: How Alpha School Made 2-Sigma RealAnime : Nathan LabenzMacKenzie Price explains how Alpha School uses adaptive apps—not chatbots—to finish core academics in two hours a day and still beat the growth models.Produit et UXAI in EducationProduct Strategy+3
5 diapositives24:25EN
Matthew Berman· 24 juin 2025Building a Sleep Tracker With Pythagora 2.0Matthew Berman · Forward FutureMatthew Berman demonstrates an AI-assisted development workflow, from an editable specification to testing and debugging a sleep-tracking prototype.Agents de codeWorkflows de développement et testsAI in healthcare+2
Diapositives1:39EN
Matthew Berman· 23 juin 2025Reddit v. Anthropic: Scraping Allegations and the Lawsuit’s ProgressMatthew Berman · Forward FutureMatthew Berman reads from Reddit’s complaint about alleged unauthorized data use; the claims remain disputed as the case proceeds.AI RegulationPrivacy & data governanceData quality & curation+1
Charger plus24 sur 1 562 affichés