13 New AI Tools I’m Actually Using

Here's a detailed summary of the YouTube video transcript:

Key Concepts

• Whisper Flow: AI-powered speech-to-text tool that cleans up dictation.
• Granola: AI meeting assistant that uses system audio and integrates notes.
• Comet AI Browser (Perplexity): AI-native browser with features like video summarization and website interaction.
• NADN: Platform for building AI agents and automations.
• Hostinger: VPS provider with a one-click N8N installation template.
• AI Creative Tools Aggregators: Platforms like Freepick, Creo, Leonardo, Art List, and Replicate that consolidate various AI image and video models.
• Magnific: AI upscaling tool for images, offering precision and creative upscaling.
• Astra (Topaz): Creative upscaler for AI video.
• Nano Banana (Google): AI image editing tool for consistency and custom graphics.
• NotebookLM (Google): AI tool for organizing and interacting with research sources.
• AI App/Mini-App Builders: Features within LLMs (Gemini, ChatGPT, Claude) for creating automated workflows.
• AI Studio (Google): Platform for building AI applications with features like "vibe coding."
• Retrieval Augmented Generation (RAG): A technique that grounds LLMs in specific data to reduce hallucinations.
• Vector Databases: Specialized databases for storing and retrieving information based on meaning.

Productivity Tools

Whisper Flow

• Functionality: Whisper Flow is an AI-powered speech-to-text tool that integrates across all devices and applications. Unlike standard dictation, it uses AI to analyze and automatically clean up spoken words.
• Key Features:
  • Intelligent Correction: Automatically fixes misspoken words, excessive "ums," and mid-sentence changes.
  • Example: The transcript provides an example where "Hey, John. Wait, who am I talking to? Sarah, could you send over the uh documents? No, no, no, not documents. videos. Could you send over the videos?" was corrected to "Hey, Sarah, could you send over the videos?"
• Use Cases: Texting, emailing, prompting ChatGPT, brainstorming ideas, dictating instructions while walking.
• Pricing: Offers a free plan with a limited word count per week, allowing users to test its capabilities.
• Impact: Described as "massive for productivity" and highly valued by the speaker.

Granola

• Functionality: An AI meeting assistant that addresses the growing annoyance of bots joining meetings.
• Key Features:
  • No Bot Required: Uses system audio to create meeting transcripts, eliminating the need for a bot to join.
  • AI-Powered Notepad: Allows users to scribble notes during calls, which are then merged with the transcript by AI for a structured output.
  • Contextual Chat: Enables users to ask questions about meeting content, such as "What did Sarah say about the budget during our meeting?" and retrieve direct quotes.
• Pricing: Offers a free basic plan, with paid options for unlimited notes and advanced integrations.

Comet AI Browser (Perplexity)

• Functionality: An AI-native browser designed to be futuristic and time-saving.
• Key Use Cases:
  • YouTube Analysis:
    • Summarize videos, extract key takeaways.
    • Chat with the video's content.
    • Specific Example: Extracting a full list of AI tools mentioned in a video with timestamps, even when chapters are vaguely named (e.g., "tool one"). This is achieved by asking Comet to "give me the full list of tools covered in this video with timestamps."
  • Website Interaction: Acts as a custom chatbot for any website.
    • Specific Example: Analyzing a course platform's landing page to assess its value proposition, layout, and suggest improvements.
  • Email Management: Can unsubscribe from emails in bulk.
    • Specific Example: The browser navigates to the promotions tab, opens emails, and unsubscribes from them, running in the background.
• Impact: Described as a "giant hassle to switch browsers" but ultimately very useful.

AI Agents and Automations

NADN

• Functionality: A powerful platform for building AI agents and automations.
• Business Use Cases:
  • Newsletter Creation: Content from YouTube channels and course platforms is fed into NADN to extract key information and reformat it into newsletters. Human review is still involved.
• Capabilities: Extremely powerful and customizable for customer service, sales, content creation, logistics, and agentic automations.
• Learning Resources: A full NADN crash course is available on the channel for in-depth learning.
• Value: Learning agents and automations is highlighted as a valuable AI skill.

Hostinger (for Self-Hosting NADN)

• Context: NADN can be set up via cloud (paid) or self-hosted. Self-hosting offers more control and lower costs.
• Self-Hosting Options:
  • Local Install (Docker): Technical, requires significant configuration.
  • Virtual Private Server (VPS): A more accessible middle ground.
• Hostinger's Advantage: Offers an easy VPS option with a one-click installation template for NADN, simplifying setup significantly.
• Setup Process (Hostinger):
  1. Navigate to Services > VPS.
  2. Select the self-hosted NADN option.
  3. Choose a plan (KVM2 recommended).
  4. Apply coupon code "Futuredia" for a discount.
  5. Complete billing and setup.
  6. Access the dashboard and manage the app.
  7. Skip initial screens, obtain a free license key from email.
  8. Enter the activation key in the NADN setup.
• Outcome: Users get a self-hosted version of NADN, described as a useful and time-saving platform.

Creative AI Tools

AI Creative Tools Aggregators

• Challenge: AI video and image tools are constantly updated, making it difficult for one tool to remain the leader.
• Solution: Using platforms that aggregate APIs from various leading tools. This avoids multiple subscriptions and the hassle of managing them.
• Examples:
  • Freepick: Praised for easy access to different models and a clean interface.
  • Creo & Leonardo: Offer unique features and tools.
  • Art List: Formerly a stock media platform, now includes AI image and video models, replacing a separate subscription for users who also use stock media.
• Recommendation: Having one platform for all models is convenient.

Replicate

• Functionality: A platform with a vast library of AI tools across various use cases, including image, video, LLMs, audio, music, and vision.
• Key Features:
  • Extensive Model Library: Thousands of models searchable and usable.
  • Pay-by-Usage: No subscription fees; users pay for what they use.
  • Connected to GitHub: Facilitates integration.
• Example Use Case: MM Audio:
  • Functionality: Generates matching sound effects for videos that lack sound.
  • Process: Upload a video without sound, click "run."
  • Outcome: Generates sound effects that match the video's feel, though not always perfectly synced. Useful for quick social media videos or B-roll.

Mirell

• Functionality: Similar to MM Audio but reported to be much better in initial tests.
• Process:
  1. Choose a video.
  2. Select "automatic."
  3. The tool analyzes the video and generates sound effects and music on separate tracks.
• Key Features:
  • Separate Tracks: Sound effects and music are on distinct tracks, allowing independent volume control.
  • Multiple Options: Generates several options for each track.
• Impact: Adds significant depth to videos and is expected to be used extensively.

Magnific

• Functionality: A suite of AI upscaling tools for images.
• Problem Solved: Addresses low resolution, errors, and warping in AI-generated images.
• Key Features:
  • Precision Upscaling: Traditional upscaling that stays close to the original.
  • Creative Upscaling: Upscales while reimagining details, adding significant detail.
  • Custom Controls: Allows users to guide the generation to be more or less like the original.
• Pricing: Described as "on the expensive side" but superior to competitors.
• Value: Extremely useful for enhancing AI image quality.

Astra (Topaz)

• Functionality: A creative upscaler specifically for AI video.
• Problem Solved: Addresses low fidelity and warping issues in AI video.
• Capabilities: Cleans up and upscales AI video significantly.
• Features: Also offers precision upscaling.
• Pricing: Expensive, but highly valuable for AI video work.

Nano Banana (Google)

• Functionality: An AI image editing tool.
• Staple Use Cases:
  • Thumbnails.
  • Consistent characters and scenes in AI videos.
  • Custom graphics.
• Process: Upload an image and request desired changes.
• Future: Nano Banana 2 is anticipated to improve upon the first version.
• Comparison: Previously won most comparisons against other AI image editing tools.
• Integration: Often combined with Magnific.

NotebookLM (Google)

• Functionality: An AI tool for organizing and interacting with learning materials.
• Key Features:
  • Multi-Source Uploads: Supports PDFs, text files, YouTube videos, and websites in a single notebook.
  • Discover Sources: Can add related material to expand context.
  • Grounded Answers: Uses uploaded sources as context, significantly reducing hallucinations by citing information sources.
  • Output Formats: Can transform research into quizzes, video overviews, reports, mind maps, flashcards, or a "podcast mode."
  • Podcast Mode: Creates a natural conversation between two AI hosts based on notes, allowing real-time questions.
• Impact: Considered the "most underrated AI tool" for information retention, allowing users to see, hear, and interact with content.
• Related Concept: Retrieval Augmented Generation (RAG) is mentioned as a core component of how NotebookLM grounds its responses.
• Vector Databases: Explained as specialized "address books for meaning" that are core to RAG.

Building Mini-Apps and Automations with LLMs

• Concept: Creating small automated workflows or "mini-apps" using LLMs like Gemini, ChatGPT, and Claude.
• Process (Gemini Example):
  1. Select the "canvas" feature.
  2. Describe the desired app in a few sentences (e.g., "title exploder combined with a thumbnail generator").
  3. The LLM generates the app.
  4. Example Workflow:
     • Input an idea.
     • Generate 20 potential titles.
     • Select a title.
     • Upload face and style reference.
     • Generate a thumbnail.
     • Edit the thumbnail using natural language.
• Development Time: This process can take only a few minutes.
• Other LLMs: Similar features exist in ChatGPT (Projects), Claude (Skills and Artifacts).
• Mechanism: Combines a knowledge base and custom instructions to create automated workflows or mini-apps.
• Benefit: Saves time for virtually everyone.

AI Studio (Google)

• Functionality: A platform offering a suite of Google AI tools (Gemini, Nano Banana, etc.) with more customization and the ability to lower safety restrictions.
• "Vibe Coding" Feature:
  • Natural Language Building: Build full apps and games using text prompts.
  • Visual Building: Draw on a canvas, add annotation boxes, and use text to request changes. This can be easier than verbal explanations.
• Purpose: For building more advanced applications than what's possible in the LLM canvas feature, offering greater capabilities and customization.

Conclusion and Further Resources

• Futuripedia Course Platform: Offers over 1,000 lessons across 30+ AI courses, including learning paths on various AI tools and concepts. A 7-day free trial is available.
• Google AI Features Video: A separate video is recommended for those interested in diving deeper into Google's free AI tools.