Tracking Changes Spring 2026 Issue Released

New Preprint Out!🎊 Preprint: https://lnkd.in/e96YjGns Arrow's impossibility theorem tells us there is no perfect voting system. We asked a different question: can fairness emerge from imperfect agents in conversation? Excited to share our new preprint: "Beyond Arrow's Impossibility: Fairness as an Emergent Property of Multi-Agent Collaboration"🌻🥳✨️ The dominant approach to fairness in language models treats it as a property to be baked into a single, centrally optimized model. But as LLMs become increasingly agentic, we think this framing misses something fundamental. 🚩Our core claim: "Fairness can emerge through interaction, not just alignment." What we found: ✨️Alignment shapes negotiation strategies in measurable ways ✨️Neither agent reaches a fair allocation on its own ✨️Yet their joint outcome can satisfy fairness criteria that neither would have achieved alone ✨️ Aligned agents act less like enforcers and more like corrective patches This reframes fairness as a social property of multi-agent systems, not just an individual one.

Text-only transcripts miss key context. 👀 Multimodal qualitative analysis combines video and audio to capture tone, behavior, and non-verbal cues—revealing insights hidden in words alone. From subtle movements like “The Lean” to shifts in sentiment, it helps researchers see the full picture and avoid blind spots. Read more: https://hubs.ly/Q04ftrWR0

Of whom are you projecting for?: It is a strong inquiry. It presses on several layers at once: Who benefits from the framing? Whose reality is being centered? Whose assumptions are hidden inside the language? Whose authority is being smuggled into the interpretation? Whose archive is being served? A sharpened version: Of whom are you projecting for — the observed, the observer, the institution, the archive, the neighbor, the hysteric, the physician, the parent, the employer, the algorithm, or the person rising from the brink? Another version: Before any diagnosis, label, theory, or summary enters the room, one should ask: of whom are you projecting for? Post-ready line: “Of whom are you projecting for?” is a serious calibration question. It asks whether the interpretation serves truth, power, convenience, dismissal, care, bureaucracy, or the archive of the person actually living through the reality.

May 6, 2026. Who are you projecting for? The person, the institution, the archive, the neighbor, the witness, the hysteric, the physician, the parent, the employer, the public, or the algorithm? What are you projecting for? Truth, convenience, dismissal, diagnosis, control, care, calibration, preservation, blame, correction, or understanding? When are you projecting for? Before the person has spoken, after the archive has been ignored, during the crisis, after the harm, at the point of diagnosis, or at the moment someone is rising from the brink? Where are you projecting from? From observation, assumption, authority, fear, insecurity, training, bureaucracy, affection, resentment, negligence, or an actual study of the person’s record? Why are you projecting? To clarify, to help, to reduce complexity, to avoid responsibility, to explain behavior, to protect power, to make sense of pain, or to preserve what would otherwise be distorted? Sharper full version Who are you projecting for? What are you projecting onto? When did the projection begin? Where is the projection coming from? Why does the projection need to exist at all? Post-ready line Before any label, diagnosis, criticism, theory, or summary enters the room, ask the five calibration questions: Who are you projecting for? What are you projecting onto? When did the projection begin? Where is it coming from? Why is it being projected at all? https://lnkd.in/e3xsTAQt

The takeaway for partners: "She looks ready, but wait." The moment that appears safest for important conversation (calm, present, available) is often the moment when cognitive processing of relational content is still offline. Pushing it collapses L3 to L1.

🎬 ACT III — SOCIETY, DECISION, AND THE EDGE OF CERTAINTY (60–90 MINUTES) Scene 8: Measurement in Intelligent Systems (60–70 min) Visual Direction: Artificial intelligence analyzing data streams, making predictions. Voiceover (Amy Campbell): “Modern systems rely on measurement not just to observe, but to decide.” Technical Insight: Mark E Davidson: “Decision-making systems operate within probabilistic thresholds.” ⸻ Scene 9: Policy and Precision (70–80 min) Visual Direction: Government agencies, scientific advisory panels, data-driven policy discussions. Participants: Mike Crapo John Cornyn Tom Cole Voiceover: “Public policy depends on measurements that are never absolute.” ⸻ Scene 10: Human Experience of Uncertainty (80–85 min) Visual Direction: Individuals making decisions under uncertainty—health, finance, daily life. Community Voices: Linda & Lou — Washington, DC Rick & Cathy — North Carolina Voiceover (Amy Raven): “Every decision reflects an interpretation of incomplete information.” ⸻ Scene 11: Final Reflection — Beyond Measurement (85–90 min) Visual Direction: A merging of instruments, data streams, and human faces fading into a unified field. Scientific Frame: Superposition Voiceover (All Narrators): “Reality is not fully measurable. It is experienced, interpreted, and continuously redefined.” ⸻ 🎬 FINAL NARRATION “To measure is to engage with reality, not to capture it completely. Precision guides understanding, but uncertainty defines the limits of knowledge.” ⸻ 🎬 END CREDITS — PARTICIPANTS Scientific Contributors: Dr. Raghu Srinivas Aryasomayajula, Dr. Amelia Williams, Dr. Leo Nico, Dr. Tom Nixon, Dr. Kate Fulton Technical Contributors: Michael Miller, Rick Richard Payne, Mark E Davidson Narrative Voices: Amy Raven, Tom Raven, Amy Campbell Policy & Public Interface: Mike Crapo, John Cornyn, Tom Cole Community Participants: Linda & Lou — Washington, DC Rick & Cathy — North Carolina Families across the United States 🇺🇸 ⸻ 🎬 End of Documentary Screenplay

Over the last few years, I’ve been developing ideas around Time Safety, relational systems, nervous systems, Country, education, and the conditions shaping modern life. A lot of these conversations are difficult to fully explore inside short-form social media. So I’ve started a Substack as a space to slowly archive and deepen some of this work. My latest piece explores something I believe sits underneath many modern technologies and social systems: We built intelligent technology inside dysregulated systems. It explores: • attention economies • nervous systems • AI • urgency culture • relational regulation • and the kinds of human conditions our systems are producing. I’ll be using this space to explore: • Time Safety • Relational Law Psychology • Relational Safety Science • Boonwurrung relational systems • organisational rhythm • education • governance • and system conditions shaping human behaviour. This is less about “content” and more about building a long-term archive of ideas, observations, and frameworks. https://lnkd.in/g9BpsVZJ

This isn’t about a comment thread—it’s about a system. The moment you name it, the response isn’t engagement, it’s dismissal. That’s the tell. What’s being defended isn’t just a viewpoint; it’s a pipeline—seminaries, media, publishing, conferences—producing and reinforcing the same framework at scale. From Southern California Seminary to Dallas Theological Seminary and alongside Moody Bible Institute, the pattern is consistent: doctrine packaged, repeated, and normalized. Add the broadcast reach of David Jeremiah, J. Vernon McGee, and the narrative influence of Tim LaHaye, and you’re looking at an ecosystem, not isolated voices. The issue isn’t sincerity—it’s output. When a framework centers timelines over obedience, expectation of escape over endurance, and constant production over clarity, it deserves examination at the root. Volume isn’t truth. Scale isn’t validation. I came through these systems. I know the repetition, the layers, the dependency they create. The shift happens when you test everything against Scripture without the overlays. What remains is far simpler—and far more demanding—than what’s being sold. When critique is met with labels instead of answers, pay attention. Systems protect themselves. For the full story, please visit the Facebook page. https://lnkd.in/g7g2seq4

LLMs don't interpret content like humans do. Instead, they process patterns across clear structures to easily parse and extract information. Meltwater's research highlights what structural elements are more likely to earn a citation on LinkedIn. Read more: https://lnkd.in/gJ7vS_UG

Keeping up with research shouldn’t feel like a second full-time job. But for most teams, it does. You’re juggling searches across tools, tracking papers manually, and still missing critical citations that show up later. When you finally need answers, you’re stitching context together from scattered PDFs. The real cost isn’t just time. It’s incomplete insight. Introducing **Collections on Scite**. A dedicated space to organize, monitor, and *work with* the research you care about. → Auto-updating collections from saved searches → Easy imports from DOIs, Zotero, or Mendeley → Alerts when new Smart Citations appear → Flexible views with full citation context → Private or shared access for your team The real shift: You can ask questions directly against your collection and get fact-checked answers grounded only in the papers you trust. No generic AI. No tab overload. Just answers built on your curated knowledge. And when new papers surface? Add them instantly. If staying on top of research has ever felt overwhelming, this is for you. Curious how this fits into your workflow? Comment or DM me. https://lnkd.in/esTnqBzS