<\/p>\n
AI models are getting more powerful, but the data they\u2019re trained on is getting worse, says Intuition founder Billy Luedtke. <\/p>\n
<\/p>\n
As AI systems grow more pervasive, users are increasingly running into limitations that are hard to fix. While the models improve, the underlying data these models are trained on remains the same. What is more, recursion, or AI models training on data generated by other AI, might actually make it worse. <\/p>\n
To talk about the future of AI, crypto.news spoke to Billy Luedtke, founder of Intuition, a decentralized protocol focused on bringing verifiable attribution, reputation, and data ownership to AI. Luedtke explains why the current data sets for AI are fundamentally flawed and what can be done to fix it. <\/p>\n
Crypto.news: Everyone right now is focused on AI infrastructure \u2014 GPUs, energy, data centers. Are people underestimating the importance of the trust layer in AI? Why is it important?<\/strong><\/p>\n Billy Luedtke: 100%. People are definitely underestimating it \u2014 and it matters for several reasons.<\/p>\n First, we\u2019re entering what I call a \u201cslop-in, slop-out\u201d era. AI is only as good as the data it consumes. But that data \u2014 especially from the open web \u2014 is largely polluted. It\u2019s not clean. It\u2019s not reflective of human intention. Much of it comes from gamified behavior online: likes, reviews, engagement hacks \u2014 all filtered through attention-optimized algorithms.<\/p>\n So when AI scrapes the internet, what it sees isn\u2019t a holistic picture of who we are. It\u2019s seeing people playing the platform. I don\u2019t behave the same way on Twitter as I do in real life. None of us do. We\u2019re optimizing for the algorithm \u2014 not expressing genuine thought.<\/p>\n It\u2019s recursive, too. The platforms train us, and we feed more distorted behavior back in. That creates a feedback loop \u2014 a spiral \u2014 that distorts AI\u2019s perception of humanity even more. We\u2019re not teaching it what we think; we\u2019re teaching it what we think will get likes.<\/p>\n The average user isn\u2019t Googling, comparing sources, or thinking critically. They\u2019re just asking ChatGPT or another model and taking the response at face value.<\/p>\n That\u2019s dangerous. If the model is opaque \u2014 a black box \u2014 and the company that controls it also controls what information you\u2019re shown or not shown, then that\u2019s total narrative control. It\u2019s centralized, unaccountable, and extremely powerful.<\/p>\n Imagine asking Grok for the best podcast, and the answer is whoever paid Elon the most. That\u2019s not intelligence \u2014 it\u2019s just advertising in disguise.<\/p>\n CN: So how do we fix that? How do we build systems that prioritize truth and value instead of engagement?<\/strong><\/p>\n BL: We need to flip the incentives. These systems should serve people \u2014 not institutions, not shareholders, not advertisers. That means building a new layer for the internet: identity and reputation primitives. That\u2019s what we\u2019re doing at Intuition.<\/p>\n We need verifiable attribution: who said what, when, and in what context. And we need a portable, decentralized reputation that helps determine how much we can trust any given source of data \u2014 not based on vibe, but on actual contextual track record.<\/p>\n Reddit is a perfect example. It\u2019s one of the largest sources<\/a> of training data for models. But if a user sarcastically says, \u201cJust k*** yourself,\u201d that can get scraped and show up in a model\u2019s recommendation to someone asking for medical advice.<\/p>\n That\u2019s horrifying \u2014 and it\u2019s what happens when models don\u2019t have context, attribution, or reputation weighting. We need to know: Is this person credible in medicine? Are they reputable in finance? Is this a trusted source, or just another random comment?<\/p>\n <\/p>\n CN: When you talk about attribution and reputation, this data needs to be stored somewhere. How do you think about that in terms of infrastructure \u2014 especially with issues like copyright and compensation?<\/strong><\/p>\n BL: That\u2019s exactly what we\u2019re solving at Intuition. Once you have verifiable attribution primitives, you know who created what data. That allows for tokenized ownership of knowledge \u2014 and with that, compensation.<\/p>\n So instead of your data living on Google\u2019s servers or OpenAI\u2019s APIs, it lives on a decentralized knowledge graph. Everyone owns what they contribute. When your data gets traversed or used in an AI output, you get a share of the value it generates.<\/p>\n That matters because right now we\u2019re digital serfs. We spend our most valuable resources \u2014 time, attention, and creativity \u2014 generating data that someone else monetizes. YouTube isn\u2019t valuable because it hosts videos; it\u2019s valuable because people curate it. Without likes, comments, or subscriptions, YouTube is worthless.<\/p>\n So we want a world where everyone can earn from the value they generate \u2014 even if you\u2019re not an influencer or extrovert. If you\u2019re consistently early to finding new artists, for example, your taste has value. You should be able to build a reputation around that and monetize it.<\/p>\n