> ## Content Index > Fetch the complete content index at: https://onthehrzn.berklee.edu/llms.txt > Use this file to discover other available public pages before exploring further. # How to Argue AI: Six Questions to Clarify Debate - URL: https://onthehrzn.berklee.edu/how-to-argue-ai-six-questions-to-clarify-the-debate/ - Published: 2026-10-02T02:27:00.000Z - Updated: 2026-10-08T18:50:22.000Z - Author: Adam Houghtaling - Tags: artificial intelligence, generative ai, assistive ai, copyright ***From radical abundance and medical miracles to environmental unease and existential dread, AI’s prolific rise and myriad impacts have fueled heated speculation and debate.*** --- *This is the first in a series of HRZN articles dedicated to demystifying and understanding artificial intelligence and its place within creative communities. As an emerging and evolving topic, we invite readers to* [*reach out and share*](mailto:beatl@berklee.edu) *their views and insights on this and other pieces in the series.* --- ## Introduction Conversations about AI are inevitably complex and often triggering. Given how quickly the technology is progressing and the extent to which it is impacting daily life, it becomes challenging to unpack without the influence of outsized CEO statements, [inflammatory social media](https://www.washingtonpost.com/technology/2021/10/26/facebook-angry-emoji-algorithm/?utm%5Fsource=cnn%5FEngagement+Party&utm%5Fmedium=email), and sensationalist headlines. Here, the baseline vocabulary, real-world arguments for and against, and some things to consider when trying to understand the incredibly complicated dynamic around artificial intelligence. This is not meant to sway, influence, or be a comprehensive look at the technology and its impacts, but neither does it shy from the extremes of debate in an attempt to honestly act as a foundation to rational thought regarding a complex topic not easily untangled. Your point of view, as always, is for you to compose. --- ## **1\. How Did We Get Here?** How did we arrive here? At this strange mixture of awe, anger, and disinterest? It didn't happen overnight. Long before the general public knew what a Large Language Model was, AI technology was already reshaping our culture. Technology has historically functioned in straight lines and rigid systems; a technological way of looking at the world that forces everything into a grid. A good example would be the development of modern audio tech platforms driving homogeneity in music production through the elimination of natural tempo fluctuations or human micro expressions. What some refer to as “the tyranny of the grid” then aligned with algorithmic recommendation feeds on platforms like Spotify and Netflix that have spent more than a decade changing how we discover, consume, and evaluate creative work. Meanwhile, 20-plus years of surgically designed social media interactions have optimized for raw emotional reaction and driven abbreviated attention spans. AI is no sudden departure from that trend. In many ways it’s just the next logical step. ![](https://storage.ghost.io/c/9a/8a/9a8a7ecd-106b-49ad-9d29-f0e9e57182b3/content/images/2026/09/brian-eno-digital-tools-quote.png) So while tech executives often describe AI as an incredible engine for personal agency, raw productivity, and the greatest opportunity for wealth creation since Web 2.0, future and working creators wonder how these tools will impact craft, ownership, the value of labor, and even notions as deep as what it means to be an artist, or even a human. That tension between seemingly endless potential and real-world impact is why arguing about AI has become the new default debate, even as controversy over technological impacts feel as old as recorded history itself. Socrates, as remembered in [Plato’s *Phaedrus*](https://www.gutenberg.org/files/1636/1636-h/1636-h.htm?ref=onthehrzn.berklee.edu), recounted the words of the mythical Egyptian king Thamus as he first encountered the written word handed down by the deity Theuth: > *“... for this discovery of yours will create forgetfulness in the learners' souls, because they will not use their memories… and you give your disciples not truth, but only the semblance of truth; they will be hearers of many things and will have learned nothing; they will appear to be omniscient and will generally know nothing; they will be tiresome company, having the show of wisdom without the reality.”* --- ## **2\. How Do We Define the Machine?** Before any rubber hits the road it’s critical to clarify what we mean when we say "AI." In general, AI—whether it be assistive, generative, predictive, etc.—is designed to help humans complete tasks, with outlying autonomous systems being an exception and not the rule. While assistive AI and generative AI are helpful categorizing terms, it’s worth understanding that the lines between most AI terminology can be blurry, as assistive tools can be generative and generative tools can be assistive. In general though, it’s generative AI at the locus of the debate around AI and this is true for multiple reasons. ### What is Generative AI? Generative AI refers to tools that create content based on patterns learned from massive training datasets. When you give a generative AI tool a prompt, it constructs a response by essentially guessing what each next word or note may be based primarily on the request and its training datasets. While this does cover all of generative AI—some image diffusion models, for example, generate through successive denoising steps—it offers a basic understanding of the technology. There is an illusion that generative AI is a massive leap forward and a special stripe of the technology, which largely comes from the nature of the output as it can seem uncanny to prompt a song in a sentence and end up with something that sounds radio ready a few seconds later. Also, because generative AI creates outputs everyone understands, like lyrics or a video, it’s easier to grasp and map directly onto our everyday lived experiences. It’s harder to know how you feel about billions of atomic bits, but easier to have an instantaneous reaction to an image or a song. --- ## **3\. What Is Authenticity?** The debate over AI-generated art hits three primary pressure points: 1. **What is Derivative? vs. What is Original?** Because generative models rely on statistical patterns from existing human works, critics view their creations as sophisticated rearrangement rather than true creative invention, or what some call “slop.” Proponents counter that synthesis, or the notion that the collision of disparate ideas creates the friction necessary for wholly new creation, is not just a human phenomenon, and that machine learning interacts with ideas in the same way with the potential to be just as uniquely innovative. 1. **Intellectual Property vs. Data Rights** Billions of dollars in enterprise value have been built by training models on books, visual art, journalism, and music. The major unresolved legal questions here revolve around how that training data was obtained, [substantial similarity](https://papers.ssrn.com/sol3/papers.cfm?abstract%5Fid=4832872&ref=onthehrzn.berklee.edu), and whether or not its employment constitutes a "fair use" interpretation of the law or wholesale infringement without consent or compensation. Where any individual artist stands is hard to guess, but labels and publishing groups have been moving forward in securing settlements and licensing deals, as [UMG and Udio](https://www.universalmusic.com/universal-music-group-and-udio-announce-udios-first-strategic-agreements-for-new-licensed-ai-music-creation-platform/?ref=onthehrzn.berklee.edu) announced in October 2025 and [Warner and Suno](https://www.wmg.com/news/warner-music-group-and-suno-forge-groundbreaking-partnership?ref=onthehrzn.berklee.edu) did in November 2025\. Where one person may see the ingestion of training data as coloring within the lines, another sees [industrial-scale plagiarism](https://authorsguild.org/news/ag-president-doug-preston-testifies-before-senate-committee/?ref=onthehrzn.berklee.edu). 1. **The Future of Work** There is understandable anxiety among students and working professionals that "good enough" generated works will undercut entry-level careers in copywriting, concept art, background scoring, or routine coding, while those already into their careers face the spectre of deskilling as reliance on generative AI output grows. ### **Synthesis & The Centaur Model** Yet, many working artists are actively integrating these tools into their workflows. ![The Centaur Model](https://storage.ghost.io/c/9a/8a/9a8a7ecd-106b-49ad-9d29-f0e9e57182b3/content/images/2026/09/the-centaur-model.png) If a creator prompts an AI model to collide 1920s Appalachian old-time music with 1990s shoegaze guitar textures, is the resulting fusion simple automation or a novel act of creative direction? Does it change after 10 prompts of editorial manipulation? Or 1,000 prompts? Or never? This hybrid approach is often called the centaur model, where human taste, experience, and intent guide the machine's computational work. For many, it serves as a cure for the dread of a blank page by generating mood boards, chord progressions, or code scaffolding that a human director then refines. In this model the artist’s role gradually shifts from manual executor to editor and curator. This is where the notion that taste, however you may define it, now becomes the primary artistic differentiator for creators, despite the fact that taste has pretty much always played that role. --- ## **4\. Where Do You Land on the Spectrum of AI Beliefs?** ![Spectrum of AI Beliefs](https://storage.ghost.io/c/9a/8a/9a8a7ecd-106b-49ad-9d29-f0e9e57182b3/content/images/2026/10/spectrum-ai-beliefs.png) The Spectrum of AI Beliefs Public discourse around AI often feels like a cultural turf war. Understanding the major philosophical camps and belief systems can help make sense of where you stand personally and why people argue about it so passionately: - **Effective Accelerationism (e/acc):** A hyper-online, techno-optimist philosophy championed by figures like Guillaume Verdon ("Beff Jezos") and investor Marc Andreessen (author of [”The Techno-Optimist Manifesto”](https://a16z.com/the-techno-optimist-manifesto/?ref=onthehrzn.berklee.edu)). Adherents view technological progress as an irresistible, net-positive force and oppose AI regulation and safety-driven slowdowns. Their core belief is that technological development should be accelerated at maximum speed to more quickly unlock solutions for climate change, energy scarcity, and disease. - **Singularitarians:** They anticipate the singularity, a future wherein artificial intelligence surpasses human intelligence and fundamentally reshapes civilization. Ray Kurzweil is perhaps the most vocal archetype: “There will be no distinction, post-Singularity, between human and machine or between physical and virtual reality.” - **Transhumanists:** The view that technology should be used to overcome human biological limits, such as aging, disease, and cognitive constraints, through means that include genetic engineering, neural interfaces, and AI. - **Defensive/Decentralized/Differential Acceleration (d/acc):** Proposed by Ethereum co-founder Vitalik Buterin in his November 2023 essay ["My Techno-Optimism,"](https://vitalik.eth.limo/general/2023/11/27/techno%5Foptimism.html?ref=onthehrzn.berklee.edu) written partly as a response to Andreessen's manifesto, d/acc proponents believe that technological progress is overwhelmingly good, but that it’s important to prioritize technologies that strengthen defense and keep power distributed. - **A Middle-Ground Majority:** While some polls suggest that concerns over AI are prominent, others, like [a recent Gallup poll](https://news.gallup.com/poll/712751/americans-cool-toward.aspx?ref=onthehrzn.berklee.edu), suggest the majority of U.S. adults (52%) see artificial intelligence as doing equal amounts of harm and good. Most people aren't posting manifestos or protesting data centers. They simply view AI as a possible problem as well as a pragmatic utility for drafting faster cover letters or debugging a spreadsheet at 4:30 on a Friday afternoon. - **Tech-Skeptics & Neo-Luddites:** These are critics who view generative AI primarily as a tool for corporate power concentration, data harvesting, and labor exploitation. Many artists find themselves in this camp with some turning back to obsolete technologies and friction-filled experiences. - **Existential-Risk & AI-Safety Advocates:** Anti-AI advocates with not-unfounded SkyNet anxieties who believe that unrestricted and unaligned machine super-intelligence poses a genuine threat to humanity’s survival. --- ## **5\. What Are the Arguments for AI?** While financial windfalls and maximizing corporate efficiency get a lot of airtime, it’s disingenuous to ignore the unlocking of challenging solutions to biological, mathematical, and logistical bottlenecks that AI systems have contributed to. **Scientific & Medical Breakthroughs:** From improved cancer screenings to cracking the code on [antibiotic candidates effective against drug-resistant bacteria](https://www.bbc.com/future/article/20260309-ai-is-finding-treatments-for-incurable-diseases?ref=onthehrzn.berklee.edu) or rapid discoveries towards [treatment for Alzheimer’s](https://today.usc.edu/how-ai-and-advanced-computing-are-accelerating-alzheimers-research/?ref=onthehrzn.berklee.edu), AI-assisted medical breakthroughs are among the most laudable results of the AI age. On the scientific side, AI is helping to shape [lithium-reduced battery materials](https://www.bbc.com/news/technology-67912033?ref=onthehrzn.berklee.edu) that may help bypass fragile global supply chains alongside environmental benefits including precision farming, more efficient and optimized power grids. ![](https://storage.ghost.io/c/9a/8a/9a8a7ecd-106b-49ad-9d29-f0e9e57182b3/content/images/2026/09/hassabis-quote-ai.png) **A New Age of Accessibility:** AI tools like [Yamaha's AI-Augmented "Anybody's Piano"](https://tech.hindustantimes.com/tech/news/anybodys-piano-ai-assisted-instrument-allows-disabled-musicians-to-perform-beethoven-71703341075413.html?ref=onthehrzn.berklee.edu) and [machine learning motion sensors](https://www.kcbd.com/2026/09/21/texas-tech-professor-develops-motion-based-musical-instrument-people-with-disabilities/?ref=onthehrzn.berklee.edu) are allowing individuals with physical disabilities to perform and create in new and inventive ways. **The Time-Saving Dividend:** The old rag “time is money” rings even more true in an economy where inflation is high and free time is scarce. If AI can transcribe a two-hour corporate meeting or isolate stems in an audio file instantly, it returns hours of life back to the user. In the history of technology convenience often sits heavy on the scale of adoption. **Radical Abundance and Post-Work Society:** Techno-optimists believe Artificial General Intelligence (AGI) will do almost all routine, physical, and cognitive work, managing production, logistics, and services while dropping the cost of goods and services to near zero. The end state of this re-thinking of labor will allow humans to spend their time on community, philosophy, arts, and experiential travel. **Geopolitical Imperatives:** Proponents around the globe argue that AI development is an international arms race with major powers struggling to stay at the forefront of the technology so as not to risk national security or the balance of international power. **The Democratization of Expertise:** Generative AI flattens the learning curve across highly gatekept industries. By acting as tireless coding aides or personalized assistants, it allows non-technical creators to build businesses and improve their lives, while providing learners with [hyper-personalized tutoring](https://www.nytimes.com/interactive/2026/09/29/magazine/ai-chatbots-brain-development-study.html?unlocked%5Farticle%5Fcode=1.FFE.yTN3.YttmHv%5FT8lf4&smid=url-share&ref=onthehrzn.berklee.edu). **Safety & Error Reduction:** People get tired, distracted, and overwhelmed; machines do not. AI is increasingly deployed to eliminate human errors in high-stakes environments from predictive maintenance on aircraft and supply chain logistics to monitoring ICU patient vitals. --- ## **6\. What are the Arguments Against AI?** Conversely, the skeptics aren't just Neo-Luddites clutching pearls over the death of traditional art. The bear case is rooted in tangible, present-day receipts: [strained electrical grids](https://techxplore.com/news/2026-09-ai-centers-grid-limits-hours.html?ref=onthehrzn.berklee.edu), [algorithmic discrimination](https://www.brookings.edu/articles/artificial-intelligence-and-algorithmic-exclusion/?ref=onthehrzn.berklee.edu), and what is perceived as the largest unlicensed strip-mining of human intellectual property in history. **Hypocrisy & Hubris of Tech Leaders:** You can cherry pick any number of tone deaf comments from Silicon Valley AI founders, but [Sam Altman’s now infamous 2015 quote](https://futureoflife.org/ai/sam-altman-investing-in-ai-safety-research/?ref=onthehrzn.berklee.edu) that "AI will probably most likely lead to the end of the world, but in the meantime, there’ll be great companies,” showcases both hubris and hypocrisy. Public warnings of existential extinction sit catbird to the billions they raise to build the self-same systems, as demonstrated in the [September 2026 Anthropic IPO filing](https://www.theguardian.com/technology/2026/sep/29/anthropic-warns-existential-ai-risks-humanity-ipo-document-claude?ref=onthehrzn.berklee.edu). **IP & Copyright Concerns:** Recently unsealed documents revealed what Microsoft Director of Applied Science Brent Hecht called “an astonishing theft of unprecedented proportions” and possibly [“the largest theft of labor in human history.”](https://arstechnica.com/tech-policy/2026/09/microsoft-exec-called-ai-scraping-the-largest-theft-of-labor-in-human-history/?ref=onthehrzn.berklee.edu) This kind of insider revelation and the force-feeding training data as quickly as possible into LLMs is driving artists to fight back using technical countermeasures like [data-poisoning tools](https://www.crowdstrike.com/en-us/cybersecurity-101/cyberattacks/data-poisoning/?ref=onthehrzn.berklee.edu) and [server-wide ingestion blockades](https://theconversation.com/ai-is-eating-website-traffic-websites-are-blocking-ai-and-reliable-information-is-getting-harder-to-find-289817?ref=onthehrzn.berklee.edu). **The Environmental Toll:** U.S. AI energy demand is projected to jump significantly during the next decade, directly undercutting global climate goals and driving communities to protest the incessant noise, resulting pollution, and lack of jobs that come with new data centers. **Economic Disruption & Labor Exploitation:** AI CEOs have not been shy with promises to governments and corporate buyers that they can eliminate large swaths of the human workforce, while [low-wage offshore data workers](https://www.wsj.com/tech/chatgpt-openai-content-abusive-sexually-explicit-harassment-kenya-workers-on-human-workers-cf191483?ref=onthehrzn.berklee.edu) do the traumatizing manual labor of filtering toxic content out of raw training datasets and even [piloting miracle AI inventions remotely](https://www.pcmag.com/news/this-companys-ai-was-really-just-remote-human-workers-pushing-buttons?ref=onthehrzn.berklee.edu). **The "Dead Internet" & Cultural Flattening:** As generative AI slop floods the web, the argument goes, we risk an ecosystem where bots post, comment, like, repost, and gobble up the Internet like *the Blob*, consuming everything in its path. This tidal wave of AI-generated content threatens to drown out genuine human curation and [launder the provenance](https://defined.ai/white-paper/data-laundering-and-ai-training?ref=onthehrzn.berklee.edu) of the human work at the core of the vast majority of AI outputs, turning the internet into a homogenized echo chamber devoid of authentic culture. As Patreon founder Jack Conte posited during his [widely watched 2026 SXSW speech](https://www.youtube.com/watch?v=Ue9-zkAz59A&ref=onthehrzn.berklee.edu): “Imagine you trained AI on every piece of music up until 1959, but nothing after that. Could it have created The Beatles? Could it have also made the leap to Carole King, Stevie Wonder, Prince, Nirvana, or Beyoncé without the people who poured their heart and soul into creating something new?” **Algorithmic Bias & Automated Discrimination:** Because machine learning models are trained on historical human data, they can reproduce or amplify bias. Delegating mortgage approvals, medical triaging, and resume filtering to opaque algorithms often results in mathematically laundering systemic racism and classism under the guise of "objective" software. **Surveillance, Deepfakes, and the Erosion of Truth:** From vehicle-surveillance networks like the heavily criticized Flock safety cameras to automated deepfakes and confidently hallucinated fake facts, trusting probabilistic guessing engines expands surveillance while eroding truth and a sense of a shared reality, fulfilling Socrates’ story of King Thamus and his ancient warning of granting the appearance of wisdom without its substance. --- ## **7\. How to Argue About AI** Whether at a dinner table or in a professional setting, when navigating discussions about AI you can keep the conversation grounded using these four foundational debate principles: 1. **Ground the Debate in Current Realities:** Steer clear of far-future science-fiction tropes (like rogue superintelligences or magical post-scarcity utopias). Focus on immediate, empirical impacts like labor trends, medical applications, copyright law, and energy consumption. 2. **Isolate the Specific Grievance:** Distinguish between objecting to technology as a whole versus objecting to a targeted use case or how a specific model was built. Frame arguments around data provenance and distinguishing between models trained on scraped public data versus those built on licensed or opt-in training data. 3. **Clarify the Role (Assistive Tool vs. Autonomous Agent):** Establish how the system is deployed. Is it an assistive utility (like an audio restoration plugin or spell-checker), or an autonomous agent delegating high-stakes decisions? Arguments lose clarity when a basic utility tool is treated like an existential threat, or an unmonitored agent is marketed as a harmless helper. 4. **Follow the Economic Incentives:** Large-scale model training and services require immense capital, specialized microchips, and massive data centers. Recognizing who owns and funds the underlying infrastructure helps shift the debate away from abstract marketing claims and toward realistic assessments of market dynamics. ![](https://storage.ghost.io/c/9a/8a/9a8a7ecd-106b-49ad-9d29-f0e9e57182b3/content/images/2026/09/nam-june-paik-quote.png) --- ## **8\. Living in the Grey** We live in a science fiction moment: autonomous taxis whip through cities, delivery droids roll down neighborhood sidewalks, the best deepfakes can convincingly deceive viewers, and generative AI can knock out a new song in seconds. Navigating the constantly shifting AI landscape isn’t easy, and it doesn't require choosing between uncritical techno-utopianism or blanket rejection. Many creators routinely leverage assistive AI tools to streamline tedious workflows while remaining thoughtful about what these shifts mean for the future of labor and craft. One can passionately advocate for robust ethical guardrails, environmental responsibility, and fair artist compensation while still acknowledging the genuinely helpful and transformative aspects of artificial intelligence tools. In fact, no matter where you fall on the spectrum of belief about AI, a well-rounded understanding of the technology—what it is, what it does, and what it impacts—will make you a better advocate for your own truth. --- Photo Credits Marc Andreessen: Andreessen Horowitz Stelarc: [Andy Miah](https://commons.wikimedia.org/wiki/File:Stelarc%5F3rd%5Fear%5F2011.jpg?ref=onthehrzn.berklee.edu) | [CC-BY-SA-2.0](https://commons.wikimedia.org/wiki/Category:CC-BY-SA-2.0?ref=onthehrzn.berklee.edu) will.i.am: Sterling Hampton Holly Herndon: Boris Camaca James Cameron: [Dick Thomas Johnson](https://www.flickr.com/people/31029865@N06?ref=onthehrzn.berklee.edu) | [CC-BY-2.0](https://commons.wikimedia.org/wiki/Category:CC-BY-2.0?ref=onthehrzn.berklee.edu) Paul McCartney: [Jerzy Bednarski](https://commons.wikimedia.org/wiki/User:Jerzy%5FBednarski?ref=onthehrzn.berklee.edu) | [CC-BY-SA-4.0](https://commons.wikimedia.org/wiki/Category:CC-BY-SA-4.0?ref=onthehrzn.berklee.edu) Stephen King: Shane Leonard Nick Cave: [Alterna2](https://www.flickr.com/photos/11718030@N07?ref=onthehrzn.berklee.edu) | [CC-BY-2.0](https://commons.wikimedia.org/wiki/Category:CC-BY-2.0?ref=onthehrzn.berklee.edu) Jack White: Big Hassle Geoffrey Hinton: [Cmichel67](https://commons.wikimedia.org/wiki/User:Cmichel67?ref=onthehrzn.berklee.edu) | [CC-BY-SA-4.0](https://commons.wikimedia.org/wiki/Category:CC-BY-SA-4.0?ref=onthehrzn.berklee.edu) Christopher Nolan: Featureflash