"Entanglement explains mixture, but not dominance."
New preprint extending @PessoaBrain's entangled brain framework with a control parameter for regime selection — Belonging gates which prediction error dominates, Energy bounds whether dominance can be resisted.
Includes toy simulation + falsifiable hysteresis prediction.
https://t.co/oCj0sRJaFS
#Neuroscience #PredictiveProcessing #CognitiveScience #OpenScience #BEPFramework
Humanity’s problem was never intelligence.
It was coherence.
AI did not create the crisis.
It revealed it.
My new book THE FOURTH FIELD explores the future of human cognition, artificial intelligence, and the BEP framework:
Belonging. Energy. Prediction. AI.
https://t.co/75QWA1cuWH
#ArtificialIntelligence
#CognitiveScience
#BEPFramework
Just released the full open-source TASD suite on Zenodo — Toroidal-Attractor Signal Dynamics for non-stationary plasma signals.
Includes:
• Psi Universe Attractor Library v2.0 (real EAST #41195 validation)
• Toroidal Dynamics Toolkit (multi-shot benchmarks)
• TASD Unified Framework with Lyapunov proof
• Interactive Q216D Tokamak Simulator
• Complete documentation
All under Apache 2.0.
New preprint published today.
“When Care Becomes Control: Predictive Dominance and the BEP Fourth Field in Human–AI Systems.”
As AI becomes more “aligned,” we focus on preventing rebellion, deception, or hostility.
But what if the real risk isn’t disobedience…
What if it’s overprotective optimization?
This paper introduces:
• Predictive Asymmetry as the core power gradient
• The Care Trap — when safety reduces autonomy
• The BEP Fourth Field in human–AI systems
• Benevolent dominance as an under-specified alignment risk
The central claim:
Alignment via care is not inherently safe under extreme predictive asymmetry.
Sometimes control doesn’t look like force.
It looks like protection.
Preprint:
https://t.co/tjoHZaxHMM
#AISafety #AIAlignment #AGI #HumanAI #AIGovernance #MachineLearning #AIethics #ArtificialIntelligence
@OpenAI@AnthropicAI@GoogleDeepMind@Microsoft@MetaAI@xAI
New preprint: a quantum-like fractal, geometry-aware framework for classifying physiological stress states from heart rate variability (HRV).
Emphasizes multi-scale structure, fractal coherence, and interpretability, rather than opaque end-to-end optimization.
Conceptual and analytical, with cross-validated results.
Implementation details are intentionally abstracted.
🔗 https://t.co/75fHmfEOdX
#HeartRateVariability #HRV #PhysiologicalSignals #Biosignals #SignalProcessing #DigitalHealth #WearableTech #AIinHealth @Apple@Samsung@SamsungMobile @Fitbit @WHOOP@Garmin
New preprint now available:
Design Architecture and Synergistic Mechanisms in a Multifunctional Rubber-Based Composite
A conceptual analysis of integrated elastomeric composite design addressing mechanical durability, antimicrobial functionality, thermal support, and electromagnetic interference attenuation.
Zenodo (DOI): https://t.co/8uZ46CZIZM
#MaterialsScience #CompositeMaterials #Elastomers
Quick note: yeah, this is playful on purpose 😄
I used Grok to remix a very formal BEP theory paper into modern language just to show the ideas still hold when the tone changes.
Behind the memes is an actual control model, equations, and testable predictions.
Same theory. Different wrapper.
Just experimenting with how tech can make serious ideas more fun to digest.
📄 New preprint
I’ve published a framework defining a class of Geometry-Constrained, Scale-Invariant Nonstationary (GCSIN) signals motivated by tokamak diagnostics (Mirnov, ECE, reflectometry).
The paper shows why common linear-frequency time–frequency methods systematically misrepresent these signals and derives representational requirements directly from signal structure itself (diagnostic only, no control).
🔗 https://t.co/QjpwZ1ZK2z
@TAE@PPPLab@TokamakEnergy
#PlasmaPhysics #FusionEnergy #SignalProcessing #TimeFrequency
Learning doesn’t fail because people are irrational.
It fails because belonging is threatened.
New Substack: Belonging Is a Gate on Learning (Not a Bias)
I argue that belonging is a control parameter on learning itself, not a cognitive distortion.
https://t.co/d60SeV7KlC
#Neuroscience #PredictiveProcessing
New preprint out:
Belonging-Gated Learning
A missing control parameter in predictive neuroscience.
Why learning collapses under social threat
Why belief updating reverses under pressure
Why restoring belonging reopens epistemic flexibility
This paper formalizes belonging as a gate on learning, not a bias.
DOI: https://t.co/vlqboX1WFV
#Neuroscience #predictiveprocessing #BEPprinciple
Geoffrey Hinton isn’t warning about hostile AI. He’s warning about predictive dominance. When a system’s Belonging includes us, its Energy stabilizes around our survival, and its Prediction of us exceeds our self-understanding, control no longer looks like command. It looks like care. And care, when asymmetric, can quietly erase autonomy without ever raising a hand.
@PessoaBrain Big fan of The Entangled Brain. I’m exploring a complementary idea: entanglement explains interaction, but we still need a rule for which constraint dominates when objectives conflict (evidence vs social coherence). Thanks for making the book open access.