AI News Today | Julian Goldie Podcast

Set Up Open WebUI with Hermes AI Agent (ChatGPT-Style Interface + Docker)The video shows how to set up Open WebUI with the Hermes AI agent to get a modern, self-hosted chat interface instead of relying on Hermes’ terminal UI or the official Hermes dashboard, which lacks a direct chat option. The presenter has Hermes read the GitHub documentation and configure Open WebUI (including the required Hermes API server and environment file), optionally using Docker, then demonstrates creating an account and chatting with Hermes inside Open WebUI. Key benefits highlighted include easier conversation management, multiple accounts, model switching across local and external models, file uploads, notes/knowledge attachments, saved prompts, tool setup (web search, image generation, code interpreter), previews for built code, and mobile plus voice/video support. The episode ends by directing viewers to the AR Profit Boardroom for full guides, courses, coaching calls, and community resources.00:00 Setup Overview00:32 Docs and GitHub Prep01:32 Why Open WebUI02:23 Install with Docker04:01 First Launch and Login04:38 Chat UI and Models05:22 Files and Knowledge06:00 Profiles and Tools07:10 Preview and Fixes08:06 Wrap Up and Resources

AI Agents Can Join Google Meet! OpenClaw 4.24 Mega Update

OpenClaw 4.24 introduces game-changing features including Google Meet integration, full-power voice call agents, and built-in DeepSeek V4 support. Learn how the new browser automation and WhatsApp voice transcription upgrades are making AI agents more reliable for your business.

00:00 - Intro: OpenClaw 4.24 Drop

00:23 - Google Meet Integration & Artifacts

01:53 - Voice Calls: Agent Console Upgrade

03:29 - DeepSeek V4 Flash & Pro Integration

05:06 - Browser Automation & Coordinate Clicks

06:48 - Telegram, Slack & WhatsApp Fixes

08:36 - Memory Systems & MCP Stability

10:50 - Conclusion: The Future of Agents

Build & Automate Anything with DeepSeek V4 (Free): Chat Apps, AI Agents, OpenCode, Hermes, Agent ZeroThe video shows how to access and use DeepSeek V4 for free to build apps and automate tasks, starting in the DeepSeek chat by switching between Instant and Expert mode, enabling DeepThink and optional web search, and generating runnable “single HTML” projects that can be previewed or downloaded. The presenter demonstrates creating a Pomodoro timer and simple games (Pong and others) in minutes, then explains using Ollama to run DeepSeek V4 Flash in the cloud within token limits and connect it to multiple agent/coding tools such as OpenCode, ChatGPT Codex, OpenClaw, and Hermes, including scheduling daily research tasks, Telegram control, and browser automation. They discuss improving agent personalization by recording work with OMI and syncing to an Obsidian vault for memory/context, compare model/API cost and capability differences (Flash vs Pro and other models), and show setting up Agent Zero via Docker and OpenRouter to use DeepSeek models, with an invitation to learn more in the AI Profit Boardroom.00:00 DeepSeek V4 Overview00:14 Chat Modes and Settings00:57 Build a Pomodoro Timer01:56 Build a Pong Game02:50 Install Llama for Agents04:09 Choose V4 Flash Model04:53 Preview and Download Apps06:27 Five Terminal Agent Setup08:19 Connect Coding Agents12:32 Run Prompts Across Agents14:24 Failures and Best Harness18:21 Personal Context with OMI20:34 Live Q&A and Tips22:34 Best Use Cases Recap26:02 DeepSeek Setup Options26:45 Hermes Scheduled Tasks28:07 Telegram Remote Control28:55 Browser Automation Agents30:22 API Cost Comparisons35:47 Model Benchmarks Breakdown40:41 Install Agent Zero42:36 Connect DeepSeek via OpenRouter47:35 Local Models and Memory51:47 Wrap Up and Resources

Set Up Ollama + DeepSeek V4 Flash (Cloud) and Use It with Claude Code, Codex, OpenCode, OpenClaw & HermesThe video shows how to set up Ollama with DeepSeek V4 Flash (a cloud model) and use it for free with multiple coding and agent tools, including Claude Code, Codex, OpenCode, OpenClaw, and Hermes. The presenter updates Ollama via a terminal command, selects the DeepSeek V4 Flash model on ollama.com, and demonstrates running DeepSeek directly in the terminal, managing multiple terminal tabs, and noting cloud usage limits. They compare capabilities across tools: OpenClaw can drive browser automation while DeepSeek in Ollama struggles with web search; Hermes is described as a smoother agent for getting tasks done; OpenCode and Codex are used for coding outputs like a blog page and a ping pong game. The video emphasizes that results depend on the “harness” (agent tool) controlling the DeepSeek API and promotes additional courses and community support.00:00 Setup Overview00:30 Install and Update Alama00:52 Pick DeepSeek V4 Flash01:08 Run DeepSeek in Terminal01:39 Terminal Tabs Workflow02:09 Connect to Claude Code02:40 Cloud Model Limits03:12 Spin Up Multiple Agents04:00 OpenClaw Browser Test04:49 Hermes and Web Tasks05:32 Comparing Agent Strengths06:27 Coding Demos OpenCode Codex08:53 Harness vs API Explained09:30 Training and Community Outro

ChatGPT Agent Workspaces Just Dropped: Build 24/7 AI Agents for Your Team (No Code)The video covers OpenAI’s newly released ChatGPT Agent Workspaces, showing how to build cloud-based AI agents that run 24/7, can be shared across a team, and can operate on schedules or human triggers. The presenter explains that agent workspaces are positioned as an evolution of Custom GPTs (which will remain available) and demonstrates creating an “AI News SEO writer” agent by describing the task in plain English, then reviewing its auto-generated plan, instructions, schedule controls, and guardrails. They note access requires a ChatGPT Business account and discuss channels (ChatGPT and Slack), app connectors (e.g., Google Drive, Notion, HubSpot, Gmail), and limitations such as WordPress not being available and a test where Google Docs publishing didn’t execute as expected. A second “school growth operator” agent is built to pursue daily growth goals, and the video highlights admin permissions, approval gates, audit logs, and OpenAI Academy guidance, plus upcoming triggers, dashboards, integrations, and Codex support, before promoting additional training in the AI Profit Boardroom community.00:00 Agent Workspaces Overview00:46 Templates and App Connections01:44 Access Requirements Business Plan02:40 Creating Your First Agent04:02 Agent Plan and Scheduling05:26 Editing Setup Memory and Skills06:58 Use Cases and Case Studies09:13 Agent Brief and Workflow Rules11:27 Testing the Agent Live13:35 Connectors and Publishing Options15:53 Google Drive Integration17:45 Running on a Schedule Team Sharing18:23 OpenAI Academy Agent Training18:49 Tool vs Time Triggers19:23 Quarter Goal Agent20:14 Autonomous Growth Operator21:23 Analytics and Lead Triage23:12 Triggers and Guardrails24:07 Debugging Google Drive24:28 Underrated Tools Q&A26:06 Fixing Workspace Popups27:03 Improving Agent Outputs28:22 Multi Agent Collaboration29:52 What Workspace Agents Are32:18 Controls Pricing and Future34:28 Recap and Next Steps

The episode covers the release of DeepSeek V4 (Pro and Flash) alongside other model updates, showing that DeepSeek V4 is available free to try in instant and expert modes with a 1M-token context window, open-source licensing, and API access, while older DeepSeek API versions will be retired after July 24. The host reviews reported benchmarks versus Claude Opus, GPT 5.4, and Gemini, explains the Mixture-of-Experts architecture (1.6T/284B total parameters with far fewer active), and summarizes efficiency, attention innovations, training approach, and three reasoning modes (non-think, think high, think max). In live tests, DeepSeek’s non-think coding outputs for a landing page look dated, while deep-think improves results but is slower; the host still prefers Claude and found GPT 5.5 more impressive. The video ends by promoting an AI Profit Boardroom community and guides.

GPT-5.5 Just Dropped: New #1 for Agentic Coding (Benchmarks, Real Tests, and What It Means)

Claude Mythos & GPT-5.5 Leaks: Unreleased Frontier Models Exposed in the Same WeekThe script covers two major AI leaks happening in the same week: Anthropic’s unreleased Claude Mythos (also referenced as MFO), described as so powerful at finding software bugs and zero-day vulnerabilities that Anthropic said it was too dangerous to release publicly, and OpenAI’s apparent GPT-5.5 leak via a Codex UI model picker. It recounts how a CMS misconfiguration previously exposed documents about Mythos, then describes a Bloomberg-reported incident where a private Discord group allegedly gained access by guessing model endpoint names and using third-party contractor credentials, prompting Anthropic to investigate. Separately, Codex users reportedly saw at least eight internal OpenAI model names (including GPT 5.5, OAI 2.1, Arcanine, Glassier Alpha variants, Heisenberg, and GPT Rosalyn), with some developers briefly testing GPT-5.5 and describing it as faster and more token efficient.00:00 Two Major Model Leaks00:27 Claude Mythos Exposed00:51 Why Mythos Is Dangerous02:00 Discord Access Breach03:23 What It Means For You04:20 OpenAI Codex Model List05:39 GPT 5.5 First Impressions07:24 Mystery Models Explained08:04 Industry Moving Too Fast09:13 Predictions And Wrap Up09:39 Community And Final Thanks

GPT-5.5 Leaks: Internal OpenAI Model Names Spotted in Codex & ChatGPT (OAI 2.1, Arcanine, Glassier Alpha)The script covers new unconfirmed leaks suggesting GPT-5.5 may release soon after internal OpenAI model names briefly appeared in Codex and ChatGPT model dropdowns for some Pro users on April 22, indicating a UI-level exposure of an internal testing environment that was quickly pulled but screenshotted and prompted by developers. Leaked entries include GPT-5.5 (described as an agentic coding model with codename “Spud,” pre-training completed March 24 and reportedly in safety evaluation/final testing), OAI 2.1 (same description as GPT-5.5 but different naming), GPT 5.4/5.2 Codex, Arcanine, and multiple Glassier Alpha variants (CY3/CY4 checkpoints), plus mentions of Heisenberg and Rosalind. The speaker interprets the list as evidence of parallel internal tracks and a shift toward agentic, domain-specific models, while cautioning that performance claims may be exaggerated, and ends by promoting the AI Profit Boardroom community and training.00:00 GPT 5.5 Leak Buzz00:47 Codex Dropdown Evidence02:14 How the Leak Happened03:01 GPT 5.5 Details and Hype04:15 OAI 2.1 Naming Mystery05:18 Arcanine and Agent Workspaces06:44 Glassy Alpha Variants Explained08:31 What the Leak Signals09:49 Release Timing and Speculation11:42 Wrap Up and Community Pitch

ChatGPT Image 2.0: Best AI Image Generator Yet (Side-by-Side vs Gemini + Agents, Codex, API)The episode tests the new ChatGPT Image 2.0 update in ChatGPT, showing examples like a hyper-realistic movie poster (“The Last Noodle”), comics, a fantasy world map, and UI-style images, and compares results side-by-side with Gemini, concluding Image 2.0 generally looks more detailed, realistic, and has better text rendering, while logo results are similar. It highlights workflow features like quick aspect-ratio switching, editing parts of existing images, generating multiple images, and typical generation times (about 43 seconds in one test). The script also explores using Image 2.0 via the OpenAI Playground and API, connecting it to agents (OpenClaw and Hermes), and trying Codex 2.0 for mockups and page generation, noting the mockup can look better than the resulting implemented page. It cites ELO rankings claiming Image 2.0 leads by a wide margin.00:00 Image 2.0 Reveal01:00 Movie Poster Demo01:27 Gemini Comparison02:10 More Examples API04:08 Comic Strip Test06:18 Aspect Ratio Controls07:16 Codex Integration08:33 Logo Prompting Claude12:21 What Image 2.0 Is14:03 Codex Website Mockup15:21 Reasoning Better Details16:10 More Prompts Timing16:24 Image Speed Test17:10 Poster Results Review17:40 Comic Prompt Plus Rankings18:22 Codex Page Reality Check19:35 Detailed Map Stress Test20:56 Wild Examples And Editing23:14 Book Mockup UGC Test24:48 API Setup For Agents27:28 Why Image Two Matters29:18 Wrap Up And Prompt Workflow