ChatGPT: OpenAI, Sam Altman, AI, Joe Rogan, Artificial Intelligence, Practical AI Anthropic's Cloud App Integrations and Hiring Challenges
Jan 28, 2026
Discussion of Anthropic's interactive cloud apps and integrations with tools like Slack, Canva, and Figma. Exploration of the Model Context Protocol and how agentic workflows enable multi-step automation. Coverage of agent permissions, prompt injection risks, and safety recommendations for app access. Conversation about AI outperforming applicants on technical assessments and how hiring tests may be redesigned.
AI Snips
Chapters
Transcript
Episode notes
Chatbot-As-A-Workspace
- Anthropic added interactive Cloud apps to run third-party tools like Slack, Canva and Figma inside Claude.
- Embedding visual apps into the chat reduces copy-paste and centralizes workflows for faster iteration.
MCP Is Becoming The Integration Standard
- The new app systems rely heavily on MCP (Model Context Protocol), an emerging cross-lab standard for app support.
- MCP adoption by Anthropic, OpenAI and Google helps make integrations interoperable across platforms.
Limit Agent Permissions
- Avoid granting broad access to agents and sensitive files; prefer dedicated folders and least privilege.
- Supervise agent workflows and never upload credentials, financial records, or unnecessary personal data.