Why does Gen Z hate AI?

Episode Summary

Executive Summary: The episode argues that AI is increasingly unpopular with young people and recent graduates who feel their job prospects are being undermined, even as industry leaders tout abundance. It also examines startup concentration in AI revenue, the tension between productivity and privacy in surveillance tech, and practical advice for founders to adapt by building independently and using AI selectively.

Main Topics: Backlash to AI in commencement speeches (Priority: 5/5): The hosts discuss Eric Schmidt’s University of Arizona commencement remarks and a similar UCF speech, arguing that students reacted negatively because they feel AI threatens their future jobs and was presented condescendingly rather than empathetically. Youth anxiety, labor markets, and AI pessimism (Priority: 5/5): A central theme is that recent graduates see AI as a direct threat to entry-level work, healthcare affordability, and economic stability, creating a generational sense of being ‘double-crossed’ by tech leadership. AI startup revenue concentration and duopoly risk (Priority: 5/5): The conversation shifts to data suggesting Anthropic and OpenAI capture the vast majority of AI startup revenue, raising concerns that app-layer startups are vulnerable to foundation-model providers moving downstream. Startups, layoffs, and the case for entrepreneurship (Priority: 4/5): The hosts argue that layoffs are becoming a signal of AI-driven efficiency and that starting a company is increasingly the best defense against becoming part of a permanent underclass. Privacy vs public safety in Flock surveillance cameras (Priority: 4/5): They debate Flock license-plate readers and AI-enabled security cameras, weighing crime-solving benefits against concerns about mass surveillance, abuse, retention policies, and the Stasi-like potential of centralized data. AI tools for personal knowledge management (Priority: 3/5): The discussion also covers AI gadgets and bookmarking/annotation tools (Plod, Annotated, ThinkWithMark) as ways to build a ‘backup brain’ that captures highlights, notes, and web annotations across media. Pragmatic AI usage: patents and local models (Priority: 3/5): In Q&A, Jason advises caution using AI for patents, and supports running models locally on hardware for privacy, cost control, and independence from token providers.

Key Arguments: Young graduates are not merely scared; they feel betrayed because AI is being framed as progress while simultaneously shrinking job opportunities. Eric Schmidt’s speech failed because it talked down to students instead of acknowledging the real labor-market anxiety they feel. College students have already spent years using ChatGPT, so they understand AI’s capabilities and limitations better than many executives assume. AI is creating a duopoly at the token layer, with Anthropic and OpenAI dominating startup AI revenue and potentially moving into applications. Foundation-model companies may use startup usage data to compete directly with their customers, making app-layer startups more fragile. Layoffs are increasingly viewed by investors as proof of AI-driven efficiency and capital discipline, even though they are painful for workers. Starting a company is presented as the best path to autonomy if AI and layoffs make traditional employment less secure. AI surveillance tools like Flock can improve public safety, but they need audit trails, retention limits, and local control to avoid abuse. Running models locally is likely to grow because privacy and token costs make cloud-only AI less attractive for some users and businesses.

Data Points: Podcast cadence: 3 shows a week - Mentioned in the opening banter about the show continuing through summer. Summer feature idea: Top 10 Twist all-stars - Planned summer series of returning favorite guests. Data center operations workforce growth: ~65,000 jobs by 2030 - Referenced from a University of Utah policy institute chart for the low-automation case. Technology layoffs: 100,000 jobs cut this year - Used to compare against the projected data center-related jobs. AI startup revenue share: 89% - Information chart cited that Anthropic and OpenAI generate 89% of AI startup revenue. Revenue concentration: 2 major companies dominate; 29 others share the remainder - Described as a long tail of smaller AI startups behind Anthropic and OpenAI. Revenue growth timeframe: Since July 2023 - The AI revenue chart was described as showing a dramatic ramp from July 2023 onward. Share change over 6 months: Anthropic and OpenAI are 4.5% more of total startup AI revenue than six months earlier - Used to argue the duopoly is strengthening. GLP medication cost reduction: From $2,000-$3,000/month to $150-$300/month - Jason compared past and current GLP pricing in the Roe.co ad read. Weight loss result: 40 pounds - Jason said he lost 40 pounds using fasting, rucking, sleep, and GLPs. BMI at peak: 32 - He noted he was technically obese at his peak BMI. Flock camera presence in Providence map: 177 cameras total; 155 Flock - Community-mapped surveillance count shown for Providence. Austin shooting response: 12 shootings reported; 200 officers involved - The Flock debate was anchored in a real case where suspects were found using a Flock-supported town nearby. Flock-supported arrest location: Manor - Suspects were arrested as they entered Flock-supported Manor after the Austin shootings.

Pivotal Quotes: "The future does not simply arrive. It gets built in laboratories, in dormitories, in startups, in classrooms, in legislators." — Eric Schmidt: From the commencement speech clip, emphasizing agency and participation in shaping AI. "They know that the horses left the barn and they understand that there just aren't going to be the number of jobs for them." — Jason: Jason explaining why young people react negatively to AI optimism from tech leaders. "If you are a worker for someone else, you're a cost center; but if you own your own company of any size, you're at least in control of your own destiny." — Jason: Part of the discussion about entrepreneurship as a defense against AI-driven labor instability.

Implications: The episode suggests AI is becoming a political and generational fault line: students distrust the narrative, startups face platform risk, and privacy concerns are rising. The winners may be builders who own their tools, data, and distribution.