935: Global Issues Accelerated by AI (with Solutions), feat. Stephanie Hare

Episode Summary

Executive Summary: Dr. Stephanie Hare argues that technology ethics means maximizing benefits and minimizing harms, and that technologists should adopt a Hippocratic-style ethos even if not a formal oath. She stresses regulating harmful use cases, not tools, warns that AI slop and social-media incentives degrade discourse, and says the biggest frontier challenge is climate and biodiversity, constrained by energy, water, and infrastructure realities.

Main Topics: Defining technology ethics: Hare frames technology ethics as a practical discipline: maximize benefits, minimize harms, and treat technology broadly as tools, processes, and systems—not just AI. Hippocratic oath for technologists: She supports the spirit of a non-enforceable professional oath for technologists as an ethos of care and responsibility, inspired by medicine’s post-WWII reckoning. Regulate use cases, not tools: Using the fork analogy, Hare argues regulation should target harmful actions and outcomes rather than the tool itself, so innovation is not unnecessarily stifled. AI slop and internet degradation: The conversation explores how generative AI may flood the internet and workplaces with low-quality content, worsening an already degraded online environment and pushing people back toward real-world interactions. Attention, media incentives, and social media addiction: Hare describes social platforms as engineered to addict users and argues that news and social media often amplify outrage because it is profitable and engaging. Climate, infrastructure, and AI limits: She warns that national AI ambitions collide with aging grids, water scarcity, and massive data-center demands, making climate and biodiversity the most urgent frontier issue. Research, broadcasting, and staying grounded: Hare explains how she avoids sensationalism by asking what real people care about, talking across sectors, and spending more time offline and in person to maintain perspective.

Key Arguments: Technology ethics should be broad: it applies to any tool or process, and the goal is to maximize benefits while minimizing harms. A professional oath for technologists would be valuable mainly as a cultural norm and training signal, not as a legal enforcement mechanism. The right regulatory target is the harmful use case, not the underlying tool; regulation should be clear, minimal, and enforceable. Medical ethics offers a useful precedent because medicine rebuilt trust after WWII by confronting the profession’s role in abuse and murder. The internet and social media are already harmful in many ways; AI may intensify content pollution and make people value offline, human-scale interactions again. People outside tech are usually more interested in practical questions like jobs, school, safety, and ROI than in inside-baseball tech narratives. Climate change and biodiversity loss deserve higher priority than frontier-tech hype because they affect everything, including AI’s own feasibility. AI deployment is constrained by physical infrastructure: electricity grids, water availability, land, and political will matter more than hype.

Data Points: Book publication year: 2022 - Technology Is Not Neutral was published in February 2022. Time in first tech job: 2000 - Hare says her first tech job began in 2000, near the end of the dot-com boom. Training at Accenture: 2 weeks - She recalls receiving only about two weeks of training before being pushed into systems work. Podcast episode number: 935 - The episode’s show notes are listed under Super Data Science episode 935. Summer heat waves in the UK: 4 - Hare mentions the UK experienced four heat waves that summer. Current UK electricity price context: Highest in Europe - She notes the UK has the highest electricity prices in Europe while discussing AI infrastructure stress. Philosophy in French schooling: Mandatory high school philosophy - She says French students must pass philosophy to graduate high school and often study it extensively. AI model productivity claim: 90–95% code generation - An ad read mentions Fabi’s AI models generating most code for SQL and Python workflows. Client productivity claim: Up to 50x faster - An ad read cites one Fabi client achieving up to 50 times faster exploratory data analysis. Insight speed improvement: 94% reduction - An ad read cites another client reducing time to business intelligence insights by 94%.

Pivotal Quotes: "How do we maximize the benefits and minimize the harms of anything that we are investing in, building, or using that could be described as a technology or a tool?" — Stephanie Hare: Her core definition of technology ethics. "The first is your first do no harm. Which I think is totally appropriate for technologists to embrace as well." — Stephanie Hare: Her argument for a Hippocratic-style ethos for tech professionals. "The majority of people on this planet are not involved in cryptocurrency and they’re not on Twitter." — Stephanie Hare: She explains why communicators should focus on what ordinary people actually care about.

Implications: Listeners should think less about hype and more about real-world harms, incentives, and infrastructure. For industry, the message is to build ethical guardrails, prioritize climate realities, and design tools that serve human needs rather than outrage or slop.