Raw LLM Responses

Why, Learn, Use, AI? Julia tells us about a study that informs us that most jobs that can be done by AI will be in a year or two. ‘She is reinforcing our warranted paranoia. No, she is saying why we need to learn how to use AI.’ Julia runs Firstmovers to help businesses use AI, ‘so she’s just promoting her own company?’ or is she sharing useful knowledge she learned running her own company? Gemini Prompt: Review, research, confirm facts, summarize, “video link, and finish with your opinion as an advanced AI learning model. Gemini’s Reply: This video, titled “State of AI: 85 Million Jobs Gone by 2027,” presented by Julia McCoy’s AI clone, provides a data-driven look at the current and near-future impact of artificial intelligence on the global workforce. Summary of the Video The video centers on a July 2025 Microsoft research paper titled “Measuring the Occupational Implications of Generative AI.” This study analyzed 200,000 real-world conversations between workers and Microsoft Copilot to determine an “AI Applicability Score” for various professions. Key Themes: Exposure vs. Elimination: The video stresses that high AI applicability (the ability for AI to assist in tasks) does not automatically mean a job will be deleted. It uses the “ATM Example”—when ATMs were introduced, bank teller jobs actually increased as they shifted toward higher-value relationship work. The Three-Stage Transition: 2025 (Surgical Strike): Automation hits white-collar roles like customer service (80% risk) and data entry first. 2026 (Replacement Effect): A massive reshuffling where 85 million jobs are displaced, but 97 million new roles emerge. 2027 (Wage Divide): A permanent gap forms between AI-literate workers (earning a 56% premium) and those left behind. Safe Havens: Jobs involving physical labor, personal care, and complex emotional intelligence (nursing, trades, construction) remain the least exposed to AI. Fact-Checking & Research I have cross-referenced the claims in the video with current industry reports (as of early 2026). Claim Status Source/Verification Microsoft Research Study Confirmed Microsoft published “Working with AI: Measuring the Occupational Implications of Generative AI” in July 2025. The methodology and top-ranked jobs (Translators, Historians, Writers) match the video. 85M Displaced / 97M New Jobs Nuanced These specific figures originate from the World Economic Forum’s (WEF) 2020 report. The WEF 2025 report updated these to 92M displaced vs. 170M new jobs by 2030, suggesting an even larger net gain. 56% AI Wage Premium Confirmed The PwC 2025 Global AI Jobs Barometer verified that workers with AI skills earn 56% more than those in identical roles without them, up from 25% the previous year. 76,440 Jobs Gone in 2025 Partially Verified This specific number appears in trackers monitoring AI-related layoffs (like those seen at Amazon, Salesforce, and Microsoft) during the 2025 fiscal year. Degrees Losing Value Confirmed PwC data shows that degree requirements for AI-augmented jobs fell by 7–9 percentage points between 2019 and 2025, favoring “skills-first” hiring. Fact-Checking Highlights Confirming the “ATM Analogy”: Historical data confirms that while ATMs took over cash dispensing, bank branches expanded, and the number of tellers grew because the cost of operating a branch dropped, allowing banks to open more locations. Confirming the “Skill Earthquake”: Industry data from 2025 supports the video’s claim that skills in AI-exposed roles are changing 66% faster than in other sectors. AI Learning Model Opinion As an advanced AI, I find the distinction between applicability and replacement to be the most critical takeaway. My “opinion” based on the data I process is as follows: The “Vibe Coding” Era: We are moving into a period where technical proficiency (like syntax-perfect coding) is becoming a commodity, while strategic intent and curation are becoming the premium skills. The Entry-Level Crisis: The video is correct to worry about the “automation cliff” for junior roles. If AI handles all the “grunt work” (summarizing, drafting, basic data sorting), companies may struggle to find ways to train the next generation of seniors who typically learn by doing those tasks. Human-in-the-Loop is Mandatory: While I can process millions of documents in seconds, I lack accountability and contextual ethics. The 56% wage premium for humans who can “pilot” AI isn’t just for knowing the tools; it’s for taking responsibility for the AI’s output. The video’s conclusion is sound: The goal isn’t to beat the machine, but to be the one who knows how to use it to solve more complex problems faster.