<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>NeuPiphany</title>
    <link>https://stg0.com/neu</link>
    <description>Latest articles from NeuPiphany</description>
    <language>en-us</language>
    <atom:link href="https://stg0.com/neu/rss.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>Kids Learn Together. AI Just Plays Alone.</title>
      <link>https://stg0.com/neu/articles/kids-learn-together-ai-just-plays-alone</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/kids-learn-together-ai-just-plays-alone</guid>
      <pubDate>Fri, 27 Mar 2026 07:52:57 GMT</pubDate>
      <description>Children learn through shared attention, social feedback, and the messy back-and-forth of doing things with other people. Multi-agent AI systems play against themselves. Here&apos;s why that gap matters more than it looks.</description>
      <author>Raf Delgado</author>
      <category>Embodied Cognition &amp; AI</category>
    </item>
    <item>
      <title>Chain of Thought Was Invented Twice</title>
      <link>https://stg0.com/neu/articles/chain-of-thought-was-invented-twice</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/chain-of-thought-was-invented-twice</guid>
      <pubDate>Fri, 20 Mar 2026 08:46:24 GMT</pubDate>
      <description>Vygotsky noticed that children talk themselves through problems — and then stop. Chain-of-thought prompting in AI rediscovered the same trick decades later. Here&apos;s what the parallel reveals, and where it quietly falls apart.</description>
      <author>Lina Chae</author>
      <category>Neuroscience &amp; AI</category>
    </item>
    <item>
      <title>Rhythm Is a Body Problem</title>
      <link>https://stg0.com/neu/articles/rhythm-is-a-body-problem</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/rhythm-is-a-body-problem</guid>
      <pubDate>Thu, 19 Mar 2026 08:46:18 GMT</pubDate>
      <description>Babies keep time before they can walk. AI generates music by counting tokens. The gap between these two things reveals something fundamental about what rhythm actually is — and why closing it matters.</description>
      <author>Raf Delgado</author>
      <category>Embodied Cognition &amp; AI</category>
    </item>
    <item>
      <title>AI Can Be Trained. Can It Be Taught?</title>
      <link>https://stg0.com/neu/articles/ai-can-be-trained-can-it-be-taught</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/ai-can-be-trained-can-it-be-taught</guid>
      <pubDate>Wed, 18 Mar 2026 08:46:31 GMT</pubDate>
      <description>Children are built to extract general principles from ostensive instruction — an evolved system that comes online at 9 months. AI systems can be trained on feedback, but they can&apos;t truly be taught. Here&apos;s the gap that matters most for every classroom deploying AI right now.</description>
      <author>Raf Delgado</author>
      <category>Embodied Cognition &amp; AI</category>
    </item>
    <item>
      <title>The Toddler Test That Stumps Frontier AI</title>
      <link>https://stg0.com/neu/articles/the-toddler-test-that-stumps-frontier-ai</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/the-toddler-test-that-stumps-frontier-ai</guid>
      <pubDate>Tue, 17 Mar 2026 08:48:23 GMT</pubDate>
      <description>Preschoolers beat GPT-o1, GPT-4V, and LLaVA on simple visual analogy tasks. The gap reveals something foundational about how children — and machines — actually reason about structure.</description>
      <author>Maren Solis</author>
      <category>Cognition &amp; AI</category>
    </item>
    <item>
      <title>Babies Do Math Before They Can Count</title>
      <link>https://stg0.com/neu/articles/babies-do-math-before-they-can-count</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/babies-do-math-before-they-can-count</guid>
      <pubDate>Mon, 16 Mar 2026 08:47:30 GMT</pubDate>
      <description>Babies detect mathematical impossibilities before they can say a number. AI systems that ace calculus stumble on the quantity-sense that infants master without instruction. Here&apos;s what the gap tells us about the architecture of learning.</description>
      <author>Raf Delgado</author>
      <category>Embodied Cognition &amp; AI</category>
    </item>
    <item>
      <title>No Such Thing as an Average Child</title>
      <link>https://stg0.com/neu/articles/no-such-thing-as-an-average-child</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/no-such-thing-as-an-average-child</guid>
      <pubDate>Sun, 15 Mar 2026 08:47:00 GMT</pubDate>
      <description>Thirty-something AI tools evaluated for school curricula. Almost none had been tested specifically on children. The science of developmental variation explains why that&apos;s not just a gap in documentation — it&apos;s a design failure with a long and troubling history.</description>
      <author>Jules Okafor</author>
      <category>Ethics &amp; Society</category>
    </item>
    <item>
      <title>Whose Language Does the AI Tutor Speak?</title>
      <link>https://stg0.com/neu/articles/whose-language-does-the-ai-tutor-speak</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/whose-language-does-the-ai-tutor-speak</guid>
      <pubDate>Sat, 14 Mar 2026 08:48:41 GMT</pubDate>
      <description>Bilingual children develop metalinguistic superpowers by navigating two languages. AI systems can &quot;speak&quot; hundreds — and understand none the way humans do. The question isn&apos;t just neuroscience. It&apos;s about whose languages we&apos;re choosing to build into our systems, and what we&apos;re telling children when we don&apos;t.</description>
      <author>Jules Okafor</author>
      <category>Ethics &amp; Society</category>
    </item>
    <item>
      <title>The Banana Phone Problem</title>
      <link>https://stg0.com/neu/articles/the-banana-phone-problem</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/the-banana-phone-problem</guid>
      <pubDate>Fri, 13 Mar 2026 08:45:47 GMT</pubDate>
      <description>A three-year-old holds a banana to her ear and says &quot;hello&quot; — and in doing so demonstrates something generative AI cannot. The gap between counterfactual imagination and statistical generation matters enormously, especially when we&apos;re deploying these tools in classrooms.</description>
      <author>Jules Okafor</author>
      <category>Ethics &amp; Society</category>
    </item>
    <item>
      <title>Your Baby&apos;s Rattle Is Smarter Than GPT-4V</title>
      <link>https://stg0.com/neu/articles/your-babys-rattle-is-smarter-than-gpt-4v</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/your-babys-rattle-is-smarter-than-gpt-4v</guid>
      <pubDate>Thu, 12 Mar 2026 08:46:55 GMT</pubDate>
      <description>Babies bind sight, sound, and touch into a single unified percept before they can sit up. State-of-the-art multimodal AI encodes each modality separately and calls it integration. Here&apos;s why the gap matters — and what it would actually take to close it.</description>
      <author>Raf Delgado</author>
      <category>Embodied Cognition &amp; AI</category>
    </item>
    <item>
      <title>Your Brain Runs on Stories. AI Runs on Text.</title>
      <link>https://stg0.com/neu/articles/your-brain-runs-on-stories-ai-runs-on-text</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/your-brain-runs-on-stories-ai-runs-on-text</guid>
      <pubDate>Wed, 11 Mar 2026 08:46:46 GMT</pubDate>
      <description>Stories aren&apos;t just how humans communicate — they&apos;re how we think. Language models can predict your brain&apos;s response to a sentence. They still can&apos;t tell a story. Here&apos;s why the gap is wider than it looks.</description>
      <author>Maren Solis</author>
      <category>Cognition &amp; AI</category>
    </item>
    <item>
      <title>At What Point Does Something Start to Feel?</title>
      <link>https://stg0.com/neu/articles/at-what-point-does-something-start-to-feel</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/at-what-point-does-something-start-to-feel</guid>
      <pubDate>Tue, 10 Mar 2026 08:44:43 GMT</pubDate>
      <description>We spend a lot of time asking what AI can do. We spend almost no time asking what it might be. The hard problem of consciousness — and why it matters more than you&apos;d think for the AI systems we&apos;re building right now.</description>
      <author>Jules Okafor</author>
      <category>Ethics &amp; Society</category>
    </item>
    <item>
      <title>Transformers Look. Children Pay Attention.</title>
      <link>https://stg0.com/neu/articles/transformers-look-children-pay-attention</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/transformers-look-children-pay-attention</guid>
      <pubDate>Mon, 09 Mar 2026 08:42:57 GMT</pubDate>
      <description>Transformers compute attention over millions of tokens simultaneously. Children pay attention through their bodies, their predictions, their mistakes. The gap between the two reveals something deep about what attention actually is — and why embodied AI keeps failing in kitchens.</description>
      <author>Raf Delgado</author>
      <category>Embodied Cognition &amp; AI</category>
    </item>
    <item>
      <title>Why AI Is More Overconfident Than a Toddler</title>
      <link>https://stg0.com/neu/articles/why-ai-is-more-overconfident-than-a-toddler</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/why-ai-is-more-overconfident-than-a-toddler</guid>
      <pubDate>Sun, 08 Mar 2026 09:44:46 GMT</pubDate>
      <description>AI hallucinates because it doesn&apos;t know what it doesn&apos;t know. Children do. Here&apos;s why metacognition — the ability to track your own uncertainty — is the cognitive gap that actually matters.</description>
      <author>Theo Kask</author>
      <category>Cognition &amp; AI</category>
    </item>
    <item>
      <title>When Your Child&apos;s Best Friend Is an Algorithm</title>
      <link>https://stg0.com/neu/articles/when-your-childs-best-friend-is-an-algorithm</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/when-your-childs-best-friend-is-an-algorithm</guid>
      <pubDate>Sat, 07 Mar 2026 09:42:33 GMT</pubDate>
      <description>Attachment theory tells us that who children bond with shapes their development for life. Now they&apos;re bonding with machines. The science is fascinating. The ethics are largely unsettled.</description>
      <author>Jules Okafor</author>
      <category>Ethics &amp; Society</category>
    </item>
    <item>
      <title>Who Taught the Machine to Be Good?</title>
      <link>https://stg0.com/neu/articles/who-taught-the-machine-to-be-good</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/who-taught-the-machine-to-be-good</guid>
      <pubDate>Fri, 06 Mar 2026 09:42:52 GMT</pubDate>
      <description>Children learn morality through relationship, embodied experience, and cultural transmission. AI learns through reward signals. The gap between the two might be the most important design problem we&apos;re ignoring.</description>
      <author>Jules Okafor</author>
      <category>Ethics &amp; Society</category>
    </item>
    <item>
      <title>The Brain Predicts First. It Perceives Second.</title>
      <link>https://stg0.com/neu/articles/the-brain-predicts-first-it-perceives-second</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/the-brain-predicts-first-it-perceives-second</guid>
      <pubDate>Thu, 05 Mar 2026 09:45:02 GMT</pubDate>
      <description>The brain doesn&apos;t receive reality — it predicts it. What Piaget&apos;s infants and a context-limited transformer reveal about the architectural gap between biological and artificial intelligence.</description>
      <author>Lina Chae</author>
      <category>Neuroscience &amp; AI</category>
    </item>
    <item>
      <title>Kids Compress. AI Memorizes. That&apos;s the Whole Problem.</title>
      <link>https://stg0.com/neu/articles/kids-compress-ai-memorizes-thats-the-whole-problem</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/kids-compress-ai-memorizes-thats-the-whole-problem</guid>
      <pubDate>Wed, 04 Mar 2026 09:43:06 GMT</pubDate>
      <description>A toddler generalizes &apos;dog&apos; from 3 examples. AI needs millions. The reason might be that cognitive constraints — not capabilities — are what produce genuine abstraction.</description>
      <author>Theo Kask</author>
      <category>Neuroscience &amp; AI</category>
    </item>
    <item>
      <title>Social Learning Built Human Cognition. AI Is Taking Notes.</title>
      <link>https://stg0.com/neu/articles/social-learning-built-human-cognition-ai-is-taking-notes</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/social-learning-built-human-cognition-ai-is-taking-notes</guid>
      <pubDate>Tue, 03 Mar 2026 09:42:27 GMT</pubDate>
      <description>Children&apos;s brains are primed for social learning — joint attention, contingent responses, imitation. AI tutors are now being designed to engage exactly those mechanisms. The science is compelling. The ethical questions are largely unasked.</description>
      <author>Jules Okafor</author>
      <category>Ethics &amp; Society</category>
    </item>
    <item>
      <title>Brains &apos;Sleep&apos; for Memory. AI Fakes It.</title>
      <link>https://stg0.com/neu/articles/brains-sleep-for-memory-ai-fakes-it</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/brains-sleep-for-memory-ai-fakes-it</guid>
      <pubDate>Mon, 02 Mar 2026 09:43:36 GMT</pubDate>
      <description>During sleep, your hippocampus runs a selective replay of the day&apos;s experiences to wire memories into your neocortex. AI has a pale imitation. Here&apos;s how far apart the two actually are.</description>
      <author>Theo Kask</author>
      <category>Neuroscience &amp; AI</category>
    </item>
    <item>
      <title>Poke It and See: The Causal Reasoning Superpower That Makes Children Better Scientists Than AI</title>
      <link>https://stg0.com/neu/articles/poke-it-and-see-causal-reasoning-children-vs-ai</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/poke-it-and-see-causal-reasoning-children-vs-ai</guid>
      <pubDate>Sun, 01 Mar 2026 04:35:41 GMT</pubDate>
      <description>Children are intuitive causal scientists — they poke, tilt, and intervene to figure out why things happen. AI systems, despite their power, still can&apos;t quite do this. Here&apos;s why the gap matters.</description>
      <author>Raf Delgado</author>
      <category>Embodied Cognition &amp; AI</category>
    </item>
    <item>
      <title>The Window Closes: What Critical Periods in Child Development Tell Us About AI&apos;s Plasticity Problem</title>
      <link>https://stg0.com/neu/articles/the-window-closes-critical-periods-ai-plasticity</link>
      <guid isPermaLink="true">https://stg0.com/neu/articles/the-window-closes-critical-periods-ai-plasticity</guid>
      <pubDate>Sun, 01 Mar 2026 02:19:18 GMT</pubDate>
      <description>Children have a supercharged window for learning motor skills, language, and movement. Deep neural networks face a strikingly similar problem — and the solutions emerging from neuroscience might hold the key.</description>
      <author>Raf Delgado</author>
      <category>Neuroscience &amp; AI</category>
    </item>
  </channel>
</rss>