Hard Fork AI

A deep dive into SoftBank’s massive $40 billion tie-up with a major AI company and what that means for industry competition. A look at a leaked document about a powerful new AI model and its surprising cybersecurity risks. Discussion of shifts in robotics strategy and a major smartphone assistant opening up to third-party AI integration.

In this episode, we explore Anthropic's recently released Claude Co-Work and Dispatch features, which allow Claude to control your computer's mouse, keyboard, and screen for automation. We look at the practical applications, setup process, and potential for streamlining repetitive tasks, noting both its current speed limitations and future potential as a powerful automation tool.Chapters00:00 Anthropic's Latest Releases01:55 Claude Co-Work & Dispatch Explained04:09 Setting Up and Initial Impressions06:56 Creating and Saving Skills08:24 Recommendations and Use Cases11:23 Developer vs. No-Coder Perspectives LinksGet the top 70+ AI Models for $8.99 at AI Box: ⁠⁠https://aibox.aiAI Chat YouTube Channel: https://www.youtube.com/@JaedenSchaferJoin my AI Hustle Community: https://www.skool.com/aihustle See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Breaking news about a major AI company pulling a popular video model and app sparks a discussion of product strategy and consolidation. Listeners hear why compute costs, competition, and shifting priorities toward productivity tools might have driven the decision. The conversation also covers community reactions, open-source hopes, and broader industry deals and rivals influencing the move.

They discuss paying engineers with cloud and model tokens as compensation. They cover Wall Street’s cool reaction to Nvidia’s big conference. Amazon’s Trainium Lab and vertical chip strategy get attention. WordPress.com adding autonomous AI agents for content management is explored. Jeff Bezos’s plan to raise $100 billion to buy and modernize manufacturers is highlighted.

Discussion of DoorDash paying couriers to capture real-world video and speech for AI training. Examination of Meta replacing human moderation with automated enforcement and incidents where internal agents misbehaved. Talk about AI-driven layoffs and industry reactions. Forecast that AI bots may generate more web traffic than humans by 2027 and how agents reshape site design and costs.

A look at a new desktop AI agent that acts like an operating system layer and raises privacy questions. A startup using real-time power optimization to cut data center energy spikes. A visual memory system that lets robots and wearables retain and share context. A shift in enterprise AI spending and a major cloud deal moving models into government regions.

A rundown of Google rolling out personalized Gemini features using Gmail and Photos. The Pentagon is pursuing AI alternatives after tensions with Anthropic. Mistral unveils Forage to let companies train custom models on their own data. BuzzFeed experiments with AI-driven content and faces quality concerns. US senators push for a shutdown of ByteDance's SeedDance over copyright and likeness issues.

A DIY mRNA vaccine for a dog designed with AI tools makes headlines. NVIDIA teases a next-gen AI chip that could boost inference performance. A $10B enterprise push aims to accelerate AI across industries through private equity partnerships. Reports suggest major tech workforce cuts to fund aggressive AI investments and reorganizations.

A rapid-build open-source AI agent created in 48 hours sparks viral attention and community contributions. The project shrinks massive codebases into a tiny, secure tool using container isolation. Interest from engineers at a major platform leads to a fast partnership and plans for commercial hosting and enterprise security services.

A fast look at a startup that raised big funding to let employees build AI agents without engineers. Discussion covers early enterprise adoption, why investors bet $50M, and how the company compares to Zapier and others. They also highlight a model-agnostic approach that lets firms swap underlying AI models to fit tasks.