AI News Today | Julian Goldie Podcast

Claude Oceanus Leaks are INSANE!

8 min · 7. juni 2026
episode Claude Oceanus Leaks are INSANE! cover

Description

Claude Oceanus Leak: Anthropic’s Mythos-Class Model May Go Public SoonThe script claims a new Anthropic model string, “Claude Oceanus V-1P,” appeared in Anthropic’s console on June 3 and leaked via a Chinese API proxy after red team access was resold, prompting Anthropic to pause the red team group and open an internal investigation. It shares unconfirmed rumors about pricing, speed, and a one-million-word memory, and argues Oceanus is built on the secret “Mythos” model family that Anthropic previously kept locked due to safety concerns. The narrator cites an Anthropic blog post saying Mythos-class models are expected to reach all customers in coming weeks, mentions reports of major productivity gains and Mythos performance claims, and suggests the release could accelerate long-horizon agentic automation. The episode promotes the AI Profit Boardroom and an agent OS to integrate new models when they drop.00:00 [https://www.youtube.com/watch?v=rKmvjaRSbUs] Oceanus Leak Shock00:36 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=36s] Secret Mythos Backstory01:02 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=62s] How the Leak Happened01:43 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=103s] Leaked Specs and Pricing02:19 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=139s] Mythos Going Public Soon02:59 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=179s] Why Businesses Should Care03:37 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=217s] Agent OS Pitch04:14 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=254s] Mythos vs Oceanus Explained04:48 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=288s] Early Performance Claims05:22 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=322s] Autonomous Work and Costs06:39 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=399s] Bigger Implications and Timeline07:33 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=453s] Final Call to Action08:01 [https://www.youtube.com/watch?v=rKmvjaRSbUs&t=481s] Wrap Up

Comments

0

Be the first to comment

Sign up now and become a member of the AI News Today | Julian Goldie Podcast community!

Get Started

1 month for 9 kr.

Then 99 kr. / month · Cancel anytime.

  • Podcasts kun på Podimo
  • 20 lydbogstimer pr. måned
  • Gratis podcasts

All episodes

469 episodes

episode China's GLM 5.2 is now FREE! artwork

China's GLM 5.2 is now FREE!

How to Use GLM 5.2 for Free on Zed AI (Plus Benchmarks vs Opus 4.8)The video explains how to access and use the open-source GLM 5.2 model for free through Zed AI (z.ai), noting token limits and showing how to select GLM 5.2 in chat or agent mode, enable web search, attach images, and switch between high/max and thinking/non-thinking settings. The creator describes GLM 5.2 as a frontier-level Chinese model with a million-token context window, released June 14, available as open weights on Hugging Face, and shares benchmark results (21 tasks, 8.23/10 on Goldy Bench) alongside comparisons where it can outperform Claude Opus 4.8 on UI/design tasks and beat Kimi K 2.7. Examples include generating playable 3D and open-world games, websites, and a working operating system with apps. They note Zed AI can be slow, recommend using a CLI via an Agent OS, and promote the AI Profit Boardroom community for tutorials and setup.00:00 [https://www.youtube.com/watch?v=JRhpe7b-GjQ] Free GLM 5.2 Access00:18 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=18s] Zed AI Setup Tour01:08 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=68s] Agent Mode Build Demo01:27 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=87s] Why GLM 5.2 Matters01:49 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=109s] Benchmarks and Specs02:36 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=156s] What It Can Build03:59 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=239s] Full OS Example04:47 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=287s] Best Way to Use It05:11 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=311s] Opus vs GLM Showdown06:53 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=413s] Tradeoffs and Self Hosting07:56 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=476s] GLM vs Kimi Comparison09:07 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=547s] Open Source Turning Point10:10 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=610s] Community and Tutorials Pitch11:49 [https://www.youtube.com/watch?v=JRhpe7b-GjQ&t=709s] Wrap Up

22. juni 202611 min
episode NEW Hermes Agent Blank Slate Update! artwork

NEW Hermes Agent Blank Slate Update!

Hermes Agent Blank Slate Setup Mode: Build a Faster, Update‑Proof AI AgentThe script explains Hermes Agent’s new Blank Slate Setup Mode, which lets you build an AI agent from the ground up with only the bare minimum (model, files, terminal) and manually enable tools later. It contrasts this with the old “fully loaded” setup where every agent boots with many capabilities and can become slower after updates, citing OpenClaw’s decline as an example. Hermes now offers Quick Setup, Full Setup, and Blank Slate, launched via typing “Hermes setup” in the terminal, with tools added later through “Hermes tools/skills” (e.g., skill opt-in). Benefits highlighted include faster performance (especially on weaker devices), more security, fewer tokens, predictable behavior, and an “update-proof” configuration that won’t load tools you didn’t choose. The presenter shows using separate agent profiles to compare blank slate vs full setups and promotes the Hermes Agent OS/community and coaching at aiprofitborne.com.00:00 [https://www.youtube.com/watch?v=I2F3-ND_5w0] Blank Slate Update01:16 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=76s] Profiles Side by Side02:18 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=138s] Update Proof Minimal Boot03:59 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=239s] Setup Options Explained05:54 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=354s] Why It Matters07:32 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=452s] Use Cases and Demo08:41 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=521s] Commands to Enable Tools09:23 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=563s] Control for Client Work10:39 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=639s] Key Takeaways Recap11:17 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=677s] Agent OS Offer12:44 [https://www.youtube.com/watch?v=I2F3-ND_5w0&t=764s] Wrap Up

Yesterday12 min
episode NEW Agent OS is INSANE! (Runs 3 Businesses FREE!) artwork

NEW Agent OS is INSANE! (Runs 3 Businesses FREE!)

Agent Operating System Q&A: Setup, Memory, Token Savings, Local vs Cloud, and Voice AgentsThis episode answers community questions about an Agent Operating System that keeps multiple AI agents and workflows in one place, including group chat orchestration, an idea-to-implementation pipeline, an agent Kanban, SEO and video agents, local engines, and a self-improving loop system, all supported by a shared memory layer. It contrasts the “old way” of juggling tabs, subscriptions, and lost context with the “new way” of a single, saved, synced workspace that reduces re-explaining. The host shares personal hardware and tool choices, explains that simpler setups can be best, and describes benchmarking models with a “Goldie Bench” to compare outputs. Additional topics include dictation tools, Hermes vs Claude trade-offs, ways to reduce token costs (free models/APIs/local), practical use cases, local GPU setups for high-quality generation, and options for Hermes voice agents.00:00 [https://www.youtube.com/watch?v=aJlMfiCL5-g] Agent OS Overview00:53 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=53s] Old Way vs New01:36 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=96s] Memory and Workflows02:03 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=123s] Community Q&A Begins02:44 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=164s] My Hardware Setup03:27 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=207s] Paperclip to Pipeline04:11 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=251s] Model Benchmarks04:54 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=294s] Dictation Apps05:21 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=321s] Hermes Troubleshooting07:04 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=424s] Avoiding Overwhelm07:52 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=472s] Cutting Token Costs08:50 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=530s] Agent OS Use Cases10:03 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=603s] Best Local AI Hardware11:55 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=715s] Jose Customizes Agent OS13:22 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=802s] Automating Substack13:53 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=833s] Running Agents on VPS14:29 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=869s] Hermes Voice Agents16:22 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=982s] Laptop and Messaging Setup16:56 [https://www.youtube.com/watch?v=aJlMfiCL5-g&t=1016s] Wrap Up and Join

Yesterday18 min
episode NEW NotebookLM AI Agent! artwork

NEW NotebookLM AI Agent!

NotebookLM’s New Update + My Free “Instant Research Engine” Agent OS SetupThis episode covers a new NotebookLM update with three major changes: an improved chat experience with added sources and over 100 curated software skills for deeper research, new visualization for data including PDFs, images, and Excel sheets, and an agentic research companion that helps start notebooks from loose ideas and questions. It then demonstrates an “agentic operating system” built around NotebookLM to better organize notebooks, chat with source-trained notebooks, run fast or deep agentic research with citations, and generate assets in a studio such as presentations, podcasts, videos, mind maps, infographics, and flashcards in parallel. The script argues this system offers better organization, more recent cited answers, model flexibility via MCP, and can be built for free, with an optional packaged version in the AI Profit Boardroom.00:00 [https://www.youtube.com/watch?v=2i5tVm_zBqo] NotebookLM Update Overview00:51 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=51s] Agent OS Dashboard Tour01:24 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=84s] Why Build a Layered System02:01 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=121s] Instant Research Engine Demo04:09 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=249s] Library Research Chat Studio Workflow05:25 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=325s] Generating Assets in Studio06:19 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=379s] Research Report Example07:42 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=462s] Parallel Agents Background Work08:37 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=517s] Free Setup Options09:37 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=577s] Output Quality Examples10:48 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=648s] Join AI Profit Boardroom11:41 [https://www.youtube.com/watch?v=2i5tVm_zBqo&t=701s] Community Support and Wrap Up

Yesterday12 min
episode Claude Agent OS is INSANE! 🤯 artwork

Claude Agent OS is INSANE! 🤯

I Built an Agentic AI Operating System (Voice-Controlled Jarvis + Video/SEO/Music Agents)Julian demonstrates an “agentic operating system” built with Claude that connects multiple AI CLIs, agents, and tools into a single workspace with a large “memory galaxy,” voice control, and web/computer navigation (e.g., “Jarvis” opening his site). He explains how he iterates daily via long livestream coding sessions to automate tasks like Substack and to integrate free tools such as NotebookLM for research and one-click generation of videos, podcasts, quizzes, slide decks, and more. He showcases agents that automate video creation (including multi-minute edited videos with research and B-roll), SEO workflows that grew sites from zero to 220+ daily clicks, and even a drum-and-bass music agent. He also highlights local-model voice coding, a multi-model group chat, benchmark/leaderboard sites, an auto-generated guide site, token/API tracking, and plugging in new models, then promotes getting the system and support via the AI Profit Boardroom community.00:00 [https://www.youtube.com/watch?v=aFD06AK2-LQ] Agent OS Overview00:38 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=38s] Voice Controlled Jarvis01:14 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=74s] Why Build Daily02:54 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=174s] Automating With Prompts03:17 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=197s] NotebookLM Studio Tools03:48 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=228s] Video Agent Automation04:38 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=278s] SEO Growth Proof05:32 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=332s] Music And Local Models06:43 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=403s] Benchmarks And Guides08:07 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=487s] Mission Control Workspace09:05 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=545s] Plug In Any Model09:36 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=576s] Community And Access11:13 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=673s] How To Start Today11:46 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=706s] Inside Profit Boardroom13:07 [https://www.youtube.com/watch?v=aFD06AK2-LQ&t=787s] Final Wrap Up

Yesterday13 min