What are normies using AI for?

Here's a summary of the provided transcript, focusing on key insights and applications of AI: **Key Applications of AI in Daily Life:** * **Parts Finding:** AI significantly simplifies finding specific parts for automotive repairs or other items. Users can input natural language descriptions (e.g., "gray 2015 Toyota Sienna, new buckle for the left side") and the AI will identify the correct part, draft emails to dealerships, and even handle ordering. This eliminates the tedious process of navigating complex filter menus and data sheets with numerous combinations. * **Data Distillation from Messy Sources:** Websites like "Tao" are integrating AI buttons that can analyze product options, breaking down complex data (like width, length, height for bolts or bearings) from messy seller data into easily digestible formats. This streamlines the process of finding specific items when source data is unorganized. * **Printer Troubleshooting:** AI agents can effectively troubleshoot common printer issues. Instead of manually searching for drivers or following complicated instructions on manufacturer websites, users can simply describe their problem to an AI, which can then guide them through repair steps, re-adding printers, or reinstalling drivers. * **Reviving Broken Tech (Landfill Material):** AI can help repurpose seemingly broken devices. An example given is rooting an Android tablet (a "Skylight calendar" with non-functional Wi-Fi), flashing custom firmware, and bridging Wi-Fi from another device. This transforms unusable tech into a functional, free Android tablet, demonstrating AI's ability to overcome hardware limitations with software solutions. * **Automating Bureaucracy and Form Filling:** For contractors, builders, and individuals, AI is a huge time-saver for navigating red tape, filling out forms, permits, and ensuring documents are correctly formatted before submission. It handles inconsistencies in formatting (e.g., postal codes in Canada) and can take loose information to correctly populate various fields in PDFs or online forms. * **NAS and Docker Container Management:** AI agents can manage and update Network Attached Storage (NAS) devices and Docker containers. This includes updating applications like Plex or Jellyfin, organizing files, and fixing broken apps. It simplifies technical tasks that many "tech-adjacent" users find challenging, making self-hosting more accessible. * **Home Assistant Maintenance:** Home Assistant setups often suffer from "rot" due to frequent disconnections or expired integrations. AI can be used to monitor Home Assistant daily, identify broken components, attempt fixes, or prompt the user for intervention. It also helps in customizing complex dashboards by generating YAML or TOML configurations, saving users from manual coding on their phones. * **Inbox and Calendar Management:** AI agents can process incoming emails from schools, extracurricular activities, and other sources to extract important information (e.g., gym clothes, field trips), download relevant attachments (photos, PDFs), and automatically add these events to a family calendar, inviting other family members as needed. This significantly reduces the manual effort of managing family schedules. * **Information Retrieval (e.g., VINs):** AI can quickly locate specific information buried in emails or other digital archives, such as a Vehicle Identification Number (VIN), saving users from manually searching through password managers or inboxes. While there are security trade-offs, modern LLMs are becoming very good at accurate retrieval rather than hallucinating information. * **Container Updates and Management (Synology NAS):** AI can automate the complex process of updating Docker containers on NAS devices like Synology. Users can instruct the AI to SSH into the device, update specific containers, and create backups, completing tasks in seconds that would otherwise be a "massive pain in the butt" due to manual steps and potential issues. * **Video Editing Automation:** AI is effective for automating non-creative aspects of video editing. Examples include: * Extracting specific quotes from interviews and placing them on a timeline. * Analyzing audio to automatically switch between camera angles based on who is speaking (e.g., lower voice for camera one, higher voice for camera two). * Transcribing multiple takes, identifying the best take, and arranging them on a timeline, saving significant time on the first pass. * Future potential includes generating and placing infographics, although current AI is not yet adept at making creative decisions about what to show on screen beyond just displaying spoken words. * **Home Lab Management (via Tailscale and SSH):** For those running home lab setups, AI agents can manage entire systems connected via Tailscale with SSH enabled. This allows the AI to perform diagnostic checks, fix issues, and repair components across the network, making complex home lab maintenance much easier. * **3D Print Design Assistance:** AI tools like "Nerb.dev" can assist in designing 3D prints, ensuring proper allowances, connections, and fit for multi-part designs. This is particularly useful for hobbyists or casual users who lack extensive CAD experience, enabling them to create functional designs for cases, stands, or other custom parts. AI can also optimize designs for better airflow and cooling based on user requirements. * **Software Command and Control (e.g., DaVinci Resolve, Fusion 360, Blender):** AI can act as a command interface for complex software. Users can type natural language requests to perform actions or troubleshoot issues they don't understand (e.g., "why can't I resize this in Fusion 360?"). This lowers the barrier to entry for casual users and hobbyists in fields like 3D modeling or video editing. * **Travel Logistics:** While AI is not recommended for generating generic travel itineraries (as everyone gets the same recommendations), it excels at organizing complicated travel data (flights, hotels, pickup times) and providing specific, sensible directions for complex journeys, especially when local information might not be readily available in the user's language via traditional search engines. * **Facebook Marketplace Listings:** AI agents can automate the tedious process of listing items on platforms like Facebook Marketplace. By taking photos, the AI can research pricing, generate generic descriptions (avoiding overly "fancy" language), and post the listing, making it easier for users to sell items they might otherwise be too lazy to list. **Ineffective/Problematic Uses of AI:** * **AI Staging for Real Estate Photos:** This is considered "awful" and deceptive. While it can make rooms look appealing, AI often distorts proportions, adds non-existent space, and creates unrealistic ideal lighting, misleading potential buyers. It's akin to using fisheye lenses to make rooms appear larger. However, AI mockups for brainstorming design ideas (e.g., Christmas light arrangements) are seen as acceptable, provided they are not presented as reality. * **Generating Generic Text/Automating Marketing:** Using AI to "extrapolate an email to be larger and more formal" or to automate marketing messages that lack genuine personalization is ineffective. Recipients quickly recognize AI-generated text, especially when it attempts to fake personal connection by referencing old content or unrelated tweets. This leads to "table stakes" where everyone does it, and it stops working. * **Generic Image Descriptions (e.g., Facebook Marketplace):** AI that simply describes what is visible in an image (e.g., "a green bicycle with tires in okay condition") is unhelpful. Users need AI to provide context and information *not* visible in the photo. * **Automating Job Applications/Proposals:** While tempting for job seekers, relying on AI to answer application questions leads to identical, generic responses that recruiters easily spot. This makes it harder for genuine candidates to stand out and contributes to an "onslaught" of indistinguishable applications, making hiring difficult. The key takeaway is that personal connections and referrals become even more crucial in an AI-saturated application landscape. * **Generic Travel Itineraries:** AI is not recommended for generating "what to do in this city" type travel plans, as everyone receives the same recommendations, leading to overcrowding at popular spots. Users should cultivate their own taste and opinions for travel experiences. **General Observations and Insights:** * **Standing Out:** In an age where many use AI for generic tasks, *not* using AI for certain things (like job application answers) can make an individual stand out. * **Busy Work Elimination:** A major benefit of AI is eliminating "busy work" – repetitive, tedious tasks that don't require creative human input. This frees up time for more valuable activities. * **Accessibility:** AI makes complex technical tasks (like managing NAS, Docker, or 3D design) accessible to non-experts or casual users who previously couldn't perform them. * **Cost of AI:** While many applications are relatively cheap, resource-intensive tasks like constantly scanning inboxes and parsing complex data can be expensive, as seen with the rapid rise and fall of OpenClaw's valuation. * **AI "Randomness" vs. Predictability:** Despite claims of AI randomness, AI often produces highly predictable and similar answers to common questions (e.g., Oxford comma, favorite movie) because it's fundamentally mathematical and requires deliberate "randomness" to avoid identical outputs. * **Personal Connections Remain Key:** Especially in job applications, personal connections and referrals become even more critical when AI automates much of the initial screening process.