Source: raw/The_Generation_That_Grew_Up_With_Algorithms_Just_Called_Bullshit_on_AI.md Creator: El (claimed PhD in computer science, YouTube channel host) URL: https://www.youtube.com/watch?v=yMzHGzLyqgE Platform: YouTube
A secondary-aggregation video essay that stitches together several primary surveys — a Gen Z sabotage report (44% figure), sentiment trend trackers, a ~6,000-executive multi-country survey, and a University of Pennsylvania student-newspaper editorial — into a single thesis. The author’s claim: Gen Z’s AI behavior is not resistance to technology, but resistance to a specific power structure being built with it. None of the underlying primary sources are cited with URLs in the transcript, so every statistic below is load-bearing only to the extent the transcript paraphrases it faithfully.
Key Statistics
Sentiment (Gen Z, past year):
- Excitement about AI dropped 14 points to 22%
- Hopefulness fell 9 points to 18%
- Anger rose 9 points to 31%
- Daily AI users (subset): excitement -18 points, hopefulness -11 points over the same period
- “The more they use it, the less hopeful they become” ^[inferred] — author’s gloss on the daily-user delta
Sabotage (44% of Gen Z workers):
- 44% of Gen Z workers admit they are actively sabotaging their company’s AI strategy
- Methods: entering proprietary information into public chatbots, refusing mandated AI tools, deliberately generating low-quality output, tampering with performance reviews to make AI appear less effective
Executive Survey (~6,000 senior executives, US/UK/Germany/Australia):
- 90% of executives report AI has had no impact on employment or productivity at their firms over the past 3 years
- 69% of firms are actively using AI
- 75% admit their company’s AI strategy is “more for show than a meaningful guide to outcomes”
- 73% of CEOs report stress or anxiety about their AI strategy
- 64% fear losing their jobs if they fail to lead the AI transition
- 48% call AI adoption a “massive disappointment”
- Only 29% see significant ROI from generative AI; 23% from AI agents
- 54% admit AI deployments are “tearing their company apart”
Usage Gap (executive vs. worker):
- 64% of executives use AI 2+ hours daily; only 28% of regular employees do
- ~1 in 5 executives use AI 4-5 hours daily
- 1 in 25 executives use AI 6+ hours daily
- 92% of executives say they are “actively cultivating an AI elite within their organizations”
Sabotage Motivations:
- 30% cited fear AI would take their own job
- 26% cited poorly executed company AI strategy — a quality judgment, not job anxiety
Executive Response (to sabotage / refusal):
- 77% say employees who refuse to become proficient in AI will not be considered for promotions or leadership roles
- 60% are considering layoffs for workers who refuse to adopt AI
Trust Gap:
- 80% of Gen Z trust AI more than their manager for certain work tasks
- Author’s framing: “not a vote of confidence in AI… a vote of no confidence in management” ^[inferred]
Security Compounding (cited in passing):
- 67% of executives believe their company has already suffered a data leak or breach because of employees using unapproved AI tools
- 35% of employees have entered proprietary information into public AI tools
Adoption Class Divide (cited in passing):
- AI super-users are 3x more likely to have received both a promotion and pay rise in the past year
- Super-users save ~9 hours per week using AI vs. ~2 hours for AI laggards (4.5x gap)
The University Front: Penn’s AI Problem
The University of Pennsylvania student newspaper published an editorial titled “Penn has an AI problem” opening with the line:
“AI cannot coexist with education. It can only degrade it. As technology advances and workers are replaced by machines, schools are some of the only places we have left to explore and wrestle with human thought.”
The editorial date is not given in the transcript ^[ambiguous].
Context the transcript provides:
- Penn was the first Ivy League school to launch an AI major
- Current AI program footprint: 10 undergraduate programs, 21 graduate programs, 8 doctoral programs
- 83% of Penn undergraduates admit to using AI in academic settings
- Penn launched its AI major by “gutting” their systems engineering program ^[inferred] — author’s characterization; primary source not cited
- Penn now implements location tracking to confirm students are physically present in lecture halls
- One student quoted: “Everyone in my immediate peer group is not using AI and is actively against it besides my two friends who are in computer science and are essentially mandated to use it.”
The author’s reading: the students writing against AI are writing from direct experience, not ignorance. 83% have used the tools; they are rejecting the deployment model after first-hand contact.
The Author’s Thesis
The core argument is that Gen Z’s pattern of AI behavior is not technology resistance but resistance to a power structure being built with the technology:
- Gen Z grew up watching social platforms exploit attention while marketing themselves as connection tools
- They watched gig-economy apps promise flexibility and deliver precarity
- They watched algorithmic feeds promise community and deliver addiction
- They are now watching executives deploy AI with the same framing — “this will empower everyone, free you for creative tasks” — while the architecture concentrates control at the executive layer
The supporting structural pattern:
- Executives use AI 2+ hours daily at 2.3x the rate of regular employees
- 92% of executives are “actively cultivating an AI elite”
- Executive response to refusal is threats (no promotions, possible layoffs)
- Sabotage is described as the only leverage workers have when they cannot vote on adoption but can degrade the technology’s apparent effectiveness one output at a time
The “more familiar, less hopeful” sentiment data is the author’s strongest evidence for the thesis: if resistance were ignorance-driven, daily use would increase hope. Instead it decreases it (excitement -18, hopefulness -11 among daily users). The author concludes: “The problem is not unfamiliarity. The problem is what becomes visible with familiarity.”
Editorial caveat
- Author identity not independently verified. The transcript narrator self-identifies as “El, PhD in computer science,” analyzing AI developments on a YouTube channel. The credential and channel identity are not corroborated outside the transcript.
- The video ends with a course pitch. The author plugs a paid two-tier course “for non-technical professionals where AI is explained simply, the actual mechanisms, not the hype” and asks viewers to complete a survey shaping the course. This is a commercial framing — not disqualifying, but worth flagging.
- Primary sources are referenced but not cited. The 44% sabotage figure, the ~6,000-executive survey (US/UK/Germany/Australia), the Gen Z sentiment trend data, the Penn editorial, and the super-user-vs-laggard productivity gap are all described as drawn from existing reports — but no source URLs, publisher names, sample sizes, methodology notes, or publication dates accompany them in the transcript.
- Treat aggregate numbers as load-bearing only after primary-source verification. Any future Karpathy-wiki article that re-cites a statistic from this entry should pull the underlying report directly. Several of these figures (90% no-impact executive admission, 44% Gen Z sabotage rate, 80% Gen Z trusts AI over manager) are high-stakes claims that would not survive a citation audit on the transcript alone.
- The author’s framing is editorial. Lines like “vote of no confidence in management” and “the only people required to learn the technology are the ones who understand it well enough to know why it might be a problem” are interpretive, not extracted from the underlying surveys. Marked
^[inferred]above where applied to numeric data.
Key Takeaways
- A YouTube video essay aggregates several 2025-2026 surveys to argue that 44% of Gen Z workers are actively sabotaging company AI strategy — and that the sabotage is quality-driven (26% cite poor strategy) as much as it is job-anxiety-driven (30%).
- Gen Z sentiment toward AI moved sharply negative over the past year: excitement -14 points (to 22%), hopefulness -9 points (to 18%), anger +9 points (to 31%). Daily users moved further negative than non-users.
- A ~6,000-executive multi-country survey reportedly found 90% of executives say AI has had no impact on employment or productivity at their firms in 3 years — while 69% are actively deploying it and 75% admit the strategy is “more for show.”
- Executives use AI 2+ hours/day at 2.3x the rate of regular employees (64% vs. 28%); 92% admit to “cultivating an AI elite.” The transcript frames this as deliberate stratification, not skill gap.
- Sabotage methods are simple and high-impact: feed proprietary data into public chatbots (security risk), refuse mandated tools, ship low-quality AI output unfixed, tamper with AI-performance reviews.
- Executive response is coercive: 77% will block promotions for AI-refusers; 60% are considering layoffs; 54% admit AI deployments are “tearing their company apart.”
- The University of Pennsylvania student newspaper published an editorial titled “Penn has an AI problem” arguing AI degrades education — written from a campus where 83% of undergrads admit to AI use and where the institution itself runs 10/21/8 AI programs and location-tracks lecture attendance.
- The author’s thesis treats Gen Z’s behavior as labor resistance to a power grab dressed as innovation, not technological resistance. Strongest evidence: the more Gen Z uses AI, the less hopeful they become — the opposite of an unfamiliarity-bred-fear pattern.
Open Questions
- Which survey is the 44% Gen Z sabotage figure from? Publisher, sample size, methodology, and publication date all needed for citation. The “30% job-fear / 26% poor-strategy / methods list” breakdown reads like it traces to a single report but is not named in the transcript.
- Which survey is the ~6,000-executive (US/UK/Germany/Australia) report? The 90%-no-impact, 75%-for-show, 54%-tearing-company-apart cluster is the most provocative dataset in the video; the underlying publisher and methodology need verification before any of these figures can be cited downstream.
- What is the precise publication date and full text of the Penn “Penn has an AI problem” editorial? The transcript quotes one paragraph; the rest of the editorial may contain stronger or weaker claims. Penn’s Daily Pennsylvanian archive is the likely source.
- Which sentiment-tracking survey produced the Gen Z deltas (-14 excitement, -9 hopefulness, +9 anger)? Gallup, Pew, Edelman, and the Anthropic-commissioned Morning Consult tracker have all run AI-sentiment work; the transcript does not specify.
Related
- WalkMe State of Digital Adoption 2026 — sister “execution gap” framing from a primary-research vantage (3,750-participant survey); maps the same executive-vs-employee perception gap from the adoption-platform vendor angle
- Stanford HAI AI Index 2026 — companion industry-wide adoption statistics; the canonical reference for cross-checking aggregate AI-adoption claims
- Gartner Strategic Impact of AI Agents — companion executive framing on agentic AI deployment patterns, complementing the executive-worker gap analysis here