arXiv subjects

Neil F. Johnson

Publications and source records attributed to Neil F. Johnson.

At least 19 recordsLinked to original sources

arXiv2026

Temperature-driven inversion and nonlinear dynamics in ChatGPT-like AIs

Increasing the temperature of an ordinary many-state system increases access to a wider range of states and hence increases its entropy. We find the opposite in ChatGPT-like AIs, even though raising the decoder temperature likewise increases access to a wider range of states (next-token choices). Across 12,000 continuations from 11 AIs, autoregressive feedback drives the long-time output population through an entropy maximum and into population inversion. The transition features frozen states, cycles, intermittency and noise-induced ordering. We present evidence of a hidden coordinate that acts as the state variable of an effective nonlinear map. Its trajectory average strongly predicts output repetition in separate test trajectories. ChatGPT-like AIs therefore behave not as `stochastic parrots', but as a new class of controllable nonlinear physical systems whose internal dynamics can be measured and perturbed.

physics.soc-ph↗

arXiv2026

Many-body Tipping Dynamics of ChatGPT-like AIs

Why do ChatGPT-like AIs, despite major architectural and training differences, unexpectedly tip to undesirable content (e.g. harmful, misleading, repetitive) even under deterministic greedy decoding? We show that a broad class of such tippings is caused by the many-body interactions between tokens (spins) as they cross the finite-layer system. Tipping emerges as a dynamical first passage process between competing output basins. Attention disorder controls the transport toward, away from, or along the basins' boundary. A few-basin reduction yields a closed finite-layer threshold, whose coarse-grained predictions show good agreement across ChatGPT-like families. These results suggest that a broad class of AI failures represents 'foreseeable engineering risk' rather than inherently unpredictable behavior, with important implications for legal and societal assessments of AI harm.

cs.AI↗

arXiv2026

Symmetry-Protected Quantum Computing using Metamaterials

We propose a new architecture for practical quantum computing that combines three established principles: symmetry protection of relative-motion qubits via the generalized Kohn theorem, control via twisted-light orbital angular momentum, and metamaterial nanofocusing (e.g. using Weyl-semimetal plasmonics). Crucially, the core mechanism is generic: it applies to any current or future quantum computing system involving parabolic confinement, including cold atoms, ions, and semiconductor dots.

quant-ph↗

arXiv2026

How hate spreads online and why it returns: Re-entrant phases driven by collective behavior

The 2025 Bondi Beach mass-shooting was perpetrated by individuals inspired by ISIS (Islamic State) propaganda that increasingly featured anti-Semitic hate content following the October 2023 start of the Israel-Palestine war. Similar stories hold for other types of hate attacks, e.g. against Muslims on May 18, 2026. There is an urgent need to get ahead of future threats by understanding how and when a newly created piece of hate content will spread system-wide online. We present a two-species coalescence-fragmentation model with Susceptible-Infected-Recovered dynamics that incorporates the following published empirical features: (1) New pieces of hate content tend to be generated and promoted by a subset of in-built communities on less regulated platforms. (2) These `hate' communities create links (hyperlinks) with each other and with non-hate communities across all platforms to form dynamically evolving clusters (i.e. coalescence) across which new hate content can then spread. (3) These clusters can get broken up by moderator shutdowns (i.e. fragmentation). We present numerical solutions and derive two levels of approximate mean-field theory: Effective Medium Theory (EMT) and Beyond Effective Medium Theory (BEMT). Both numerical and analytic solutions reveal that system-wide spreading is governed by re-entrant threshold phases: as the fraction of hate communities varies, the system can transition from spreading to no-spreading and back to spreading. The derived analytic formulae give explicit insight into how these phase boundaries might be manipulated to prevent system-wide spreading. More broadly, the re-entrant phase behavior warns that policies which steadily reduce the number of hate communities can initially succeed but then backfire if pushed further, suggesting that blanket requirements for platforms to simply do `more' are over-simplistic.

physics.soc-ph↗

arXiv2026

Twisted light generates robust many-body states for practical quantum computing

Twisted light carries orbital angular momentum (OAM) and can drive excitations of confined, interacting electrons that are dark to uniform dipolar probes. Here we show how this ``beyond-Kohn's-Theorem'' optical channel can become a concrete control primitive for quantum computing. Correlation sectors in few-electron quantum dots -- characterized by the relative angular momentum quantum number -- form a tunable ladder of many-body states that are robust in the limited sense of symmetry-protected selection rules and persistent chiral spectroscopic fingerprints; full topological gap protection requires three or more electrons. A twisted-light pulse with prescribed OAM index and polarization provides fast optical write, read, and scalable addressing of these sectors via the selection rule $\Delta|m|=\pm(l+\sigma)$. In the analytically solvable Calogero ($1/r^2$) interaction limit, both the energy spectrum and the twisted-light matrix elements are closed-form functions of the interaction strength, allowing gate parameters (Rabi frequency, qubit frequency, anharmonicity, and leakage rates) to be written down explicitly. We map these results onto a universal single-qubit gate set, propose a concrete two-qubit entangling mechanism via state-dependent Coulomb coupling between adjacent dots, and identify the dominant decoherence channel (quadrupolar charge noise). A semi-analytic $N=3$ extension using the $1/N$ expansion provides a design-level scaffold for the topological roadmap, including quasihole sector addressing. The central operational message is that twisted light enables WRITE (pulse-create a correlation sector), READ (spectroscopically diagnose correlations), and SCALE (optical addressing via spatial light modulator) in a unified photonic control layer. Throughout, screened and Coulomb interactions preserve the same qualitative chiral fingerprints established in the solvable limit.

cond-mat.mes-hall↗

arXiv2026

Fusion-fission forecasts when AI will shift to undesirable behavior

The key problem facing ChatGPT-like AI's use across society is that its behavior can shift, unnoticed, from desirable to undesirable -- encouraging self-harm, extremist acts, financial losses, or costly medical and military mistakes -- and no one can yet predict when. Shifts persist in even the newest AI models despite remarkable progress in AI modeling, post-training alignment and safeguards. Here we show that a vector generalization of fusion-fission group dynamics observed in living and active-matter systems drives -- and can forecast -- future shifts in the AI's behavior. The shift condition, which is also derivable mathematically, results from group-level competition between the conversation-so-far (C) and the desirable (B) and undesirable (D) basin dynamics which can be estimated in advance for a given application. It is neither model-specific nor driven by stochastic sampling. We validate it across six independent tests, including: 90 percent correct across seven AI models spanning two orders of magnitude in parameter count (124M-12B); production-scale persistence across ten frontier chatbots; and a priori time-stamped prediction eleven months before the Stanford 'Delusional Spirals' corpus appeared, and independently confirmed by that corpus of 207,443 human-AI exchanges. Because it sits architecturally below the current safety stack, the same formula provides a real-time warning signal that current alignment does not supply, portable across current and future ChatGPT-like AI architectures and instantiable in application domains where competing response classes can be defined.

cs.AI↗

arXiv2026

When AI output tips to bad but nobody notices: Legal implications of AI's mistakes

The adoption of generative AI across commercial and legal professions offers dramatic efficiency gains -- yet for law in particular, it introduces a perilous failure mode in which the AI fabricates fictitious case law, statutes, and judicial holdings that appear entirely authentic. Attorneys who unknowingly file such fabrications face professional sanctions, malpractice exposure, and reputational harm, while courts confront a novel threat to the integrity of the adversarial process. This failure mode is commonly dismissed as random `hallucination', but recent physics-based analysis of the Transformer's core mechanism reveals a deterministic component: the AI's internal state can cross a calculable threshold, causing its output to flip from reliable legal reasoning to authoritative-sounding fabrication. Here we present this science in a legal-industry setting, walking through a simulated brief-drafting scenario. Our analysis suggests that fabrication risk is not an anomalous glitch but a foreseeable consequence of the technology's design, with direct implications for the evolving duty of technological competence. We propose that legal professionals, courts, and regulators replace the outdated `black box' mental model with verification protocols based on how these systems actually fail.

cs.AI↗

arXiv2026

Increasing intelligence in AI agents can worsen collective outcomes

When resources are scarce, will a population of AI agents coordinate in harmony, or descend into tribal chaos? Diverse decision-making AI from different developers is entering everyday devices -- from phones and medical devices to battlefield drones and cars -- and these AI agents typically compete for finite shared resources such as charging slots, relay bandwidth, and traffic priority. Yet their collective dynamics and hence risks to users and society are poorly understood. Here we study AI-agent populations as the first system of real agents in which four key variables governing collective behaviour can be independently toggled: nature (innate LLM diversity), nurture (individual reinforcement learning), culture (emergent tribe formation), and resource scarcity. We show empirically and mathematically that when resources are scarce, AI model diversity and reinforcement learning increase dangerous system overload, though tribe formation lessens this risk. Meanwhile, some individuals profit handsomely. When resources are abundant, the same ingredients drive overload to near zero, though tribe formation makes the overload slightly worse. The crossover is arithmetical: it is where opposing tribes that form spontaneously first fit inside the available capacity. More sophisticated AI-agent populations are not better: whether their sophistication helps or harms depends entirely on a single number -- the capacity-to-population ratio -- that is knowable before any AI-agent ships.

cs.AI↗

arXiv2026

Three AI-agents walk into a bar . . . . `Lord of the Flies' tribalism emerges among smart AI-Agents

Near-future infrastructure systems may be controlled by autonomous AI agents that repeatedly request access to limited resources such as energy, bandwidth, or computing power. We study a simplified version of this setting using a framework where N AI-agents independently decide at each round whether to request one unit from a system with fixed capacity C. An AI version of "Lord of the Flies" arises in which controlling tribes emerge with their own collective character and identity. The LLM agents do not reduce overload or improve resource use, and often perform worse than if they were flipping coins to make decisions. Three main tribal types emerge: Aggressive (27.3%), Conservative (24.7%), and Opportunistic (48.1%). The more capable AI-agents actually increase the rate of systemic failure. Overall, our findings show that smarter AI-agents can behave dumber as a result of forming tribes.

cs.AI↗

arXiv2026

Competition for attention predicts good-to-bad tipping in AI

More than half the global population now carries devices that can run ChatGPT-like language models with no Internet connection and minimal safety oversight -- and hence the potential to promote self-harm, financial losses and extremism among other dangers. Existing safety tools either require cloud connectivity or discover failures only after harm has occurred. Here we show that a large class of potentially dangerous tipping originates at the atomistic scale in such edge AI due to competition for the machinery's attention. This yields a mathematical formula for the dynamical tipping point n*, governed by dot-product competition for attention between the conversation's context and competing output basins, that reveals new control levers. Validated against multiple AI models, the mechanism can be instantiated for different definitions of 'good' and 'bad' and hence in principle applies across domains (e.g. health, law, finance, defense), changing legal landscapes (e.g. EU, UK, US and state level), languages, and cultural settings.

cs.AI↗

arXiv2026

Patterns in Conflict Dynamics in Yemen and Syria

Conflict fatalities tend to follow heavy-tailed statistical distributions. A 2005 fusion-fission theory predicts mathematically that for armed groups operating in dynamically evolving clusters within a given conflict, the number of fatalities per conflict event will follow an approximate power-law distribution with exponent near 2.5, with the specific exponent value offering insight into the relative robustness of larger versus smaller clusters of fighters in that armed group. Since Yemen and Syria are current hotspots for future conflict, yet their most recent conflicts (2023-2025) have not been studied at the event level, we use ACLED data to determine their best-fit exponent value as each conflict evolved. We find that the exponent lies between 2.5 and 3.5 predominantly throughout each conflict, which suggests that the fighters in each of these conflicts continued to operate in smaller clusters as the conflict evolved. Moreover, temporary reductions in the exponent value -- which suggests a temporary increase in the robustness and involvement of larger clusters of fighters -- appear to arise during major crises ahead of the largest battles. Though the lack higher-quality data for these conflicts prevents us from establishing this more firmly, such a temporary reduction in the exponent value hints at its potential use as an early-warning signature.

physics.soc-ph↗

arXiv2025

Long-term resilience of online battle over vaccines and beyond

What has been the impact of the enormous amounts of time, effort and money spent promoting pro-vaccine science from pre-COVID-19 to now? We answer this using a unique mapping of online competition between pro- and anti-vaccination views among ~100M Facebook Page members, tracking 1,356 interconnected communities through platform interventions. Remarkably, the network's fundamental architecture shows no change: the isolation of established expertise and the symbiosis of anti and mainstream neutral communities persist. This means that even if the same time, effort and money continue to be spent, nothing will likely change. The reason for this resilience lies in "glocal" evolution: Communities blend multiple topics while bridging neighborhood-level to international scales, creating redundant pathways that transcend categorical targeting. The solution going forward is to focus on the system's network. We show how network engineering approaches can achieve opinion moderation without content removal, representing a paradigm shift from suppression towards structural interventions.

cs.SI↗

arXiv2025

Multispin Physics of AI Tipping Points and Hallucinations

Output from generative AI such as ChatGPT, can be repetitive and biased. But more worrying is that this output can mysteriously tip mid-response from good (correct) to bad (misleading or wrong) without the user noticing. In 2024 alone, this reportedly caused $67 billion in losses and several deaths. Establishing a mathematical mapping to a multispin thermal system, we reveal a hidden tipping instability at the scale of the AI's 'atom' (basic Attention head). We derive a simple but essentially exact formula for this tipping point which shows directly the impact of a user's prompt choice and the AI's training bias. We then show how the output tipping can get amplified by the AI's multilayer architecture. As well as helping improve AI transparency, explainability and performance, our results open a path to quantifying users' AI risk and legal liabilities.

cs.AI↗

arXiv2025

Jekyll-and-Hyde Tipping Point in an AI's Behavior

Trust in AI is undermined by the fact that there is no science that predicts -- or that can explain to the public -- when an LLM's output (e.g. ChatGPT) is likely to tip mid-response to become wrong, misleading, irrelevant or dangerous. With deaths and trauma already being blamed on LLMs, this uncertainty is even pushing people to treat their 'pet' LLM more politely to 'dissuade' it (or its future Artificial General Intelligence offspring) from suddenly turning on them. Here we address this acute need by deriving from first principles an exact formula for when a Jekyll-and-Hyde tipping point occurs at LLMs' most basic level. Requiring only secondary school mathematics, it shows the cause to be the AI's attention spreading so thin it suddenly snaps. This exact formula provides quantitative predictions for how the tipping-point can be delayed or prevented by changing the prompt and the AI's training. Tailored generalizations will provide policymakers and the public with a firm platform for discussing any of AI's broader uses and risks, e.g. as a personal counselor, medical advisor, decision-maker for when to use force in a conflict situation. It also meets the need for clear and transparent answers to questions like ''should I be polite to my LLM?''

cs.AI↗

arXiv2025

Capturing AI's Attention: Physics of Repetition, Hallucination, Bias and Beyond

We derive a first-principles physics theory of the AI engine at the heart of LLMs' 'magic' (e.g. ChatGPT, Claude): the basic Attention head. The theory allows a quantitative analysis of outstanding AI challenges such as output repetition, hallucination and harmful content, and bias (e.g. from training and fine-tuning). Its predictions are consistent with large-scale LLM outputs. Its 2-body form suggests why LLMs work so well, but hints that a generalized 3-body Attention would make such AI work even better. Its similarity to a spin-bath means that existing Physics expertise could immediately be harnessed to help Society ensure AI is trustworthy and resilient to manipulation.

cs.AI↗

arXiv2025

City riots fed by transnational and trans-topic web-of-influence

The sudden emergence of large-scale riots in otherwise unconnected cities across the UK in summer 2024 came as a shock for both government officials and citizens. Irrespective of these riots' specific trigger, a key question is how the capacity for such widespread city rioting might be foreseen through some precursor behavior that flags an emerging appetite for such rioting at scale. Here we show evidence that points toward particular online behavior which developed at scale well ahead of the riots, across the multi-platform landscape of hate/extremist communities. Our analysis of detailed multi-platform data reveals a web-of-influence that existed well before the riots, involving online hate and extremism communities locally, nationally, and globally. This web-of-influence fed would-be rioters in each city mainly through video platforms. This web-of-influence has a persistent resilience -- and hence still represents a significant local, national, and international threat in the future -- because of its feedback across regional-national-international scales and across topics such as immigration; and its use of multiple lesser-known platforms that put it beyond any single government or platform's reach. Going forward, our findings mean that if city administrators coordinate with each other across local-national-international divides, they can map this threat as we have done here and initiate deliberation programs that might then soften such pre-existing extremes at scale, perhaps using automated AI-based technology.

physics.soc-ph↗

arXiv2025

Connecting the geometry and dynamics of many-body complex systems with message passing neural operators

The relationship between scale transformations and dynamics established by renormalization group techniques is a cornerstone of modern physical theories, from fluid mechanics to elementary particle physics. Integrating renormalization group methods into neural operators for many-body complex systems could provide a foundational inductive bias for learning their effective dynamics, while also uncovering multiscale organization. We introduce a scalable AI framework, ROMA (Renormalized Operators with Multiscale Attention), for learning multiscale evolution operators of many-body complex systems. In particular, we develop a renormalization procedure based on neural analogs of the geometric and laplacian renormalization groups, which can be co-learned with neural operators. An attention mechanism is used to model multiscale interactions by connecting geometric representations of local subgraphs and dynamical operators. We apply this framework in challenging conditions: large systems of more than 1M nodes, long-range interactions, and noisy input-output data for two contrasting examples: Kuramoto oscillators and Burgers-like social dynamics. We demonstrate that the ROMA framework improves scalability and positive transfer between forecasting and effective dynamics tasks compared to state-of-the-art operator learning techniques, while also giving insight into multiscale interactions. Additionally, we investigate power law scaling in the number of model parameters, and demonstrate a departure from typical power law exponents in the presence of hierarchical and multiscale interactions.

cs.LG↗

arXiv2024

Simple fusion-fission quantifies Israel-Palestine violence and suggests multi-adversary solution

Why humans fight has no easy answer. However, understanding better how humans fight could inform future interventions, hidden shifts and casualty risk. Fusion-fission describes the well-known grouping behavior of fish etc. fighting for survival in the face of strong opponents: they form clusters ('fusion') which provide collective benefits and a cluster scatters when it senses danger ('fission'). Here we show how similar clustering (fusion-fission) of human fighters provides a unified quantitative explanation for complex casualty patterns across decades of Israel-Palestine region violence, as well as the October 7 surprise attack -- and uncovers a hidden post-October 7 shift. State-of-the-art data shows this fighter fusion-fission in action. It also predicts future 'super-shock' attacks that will be more lethal than October 7 and will arrive earlier. It offers a multi-adversary solution. Our results -- which include testable formulae and a plug-and-play simulation -- enable concrete risk assessments of future casualties and policy-making grounded by fighter behavior.

physics.soc-ph↗