Ahead of the hit show's finale, cocreators Paul W. Downs and Lucia Aniello talk about media consolidation, the perils of censorship, and why they find AI "deeply disturbing." If you're a WIRED reader who uses AI in any creative context, I'd suggest staying far, far away from anyone involved in the TV show . In an interview earlier this year, actor Hannah Einbinder (who plays young comedy writer Ava Daniels on the show) described AI creators as "losers," "not artists," and "not special." In a wide-ranging conversation for ahead of the series finale on HBO Max, Paul W. Downs and Lucia Aniello were resolute about the value of human creativity--and what can be lost when AI enters the picture. If their work on is any indication, Downs and Aniello (along with their third cocreator, Jen Statsky) would be wise to stick with the tough, tiring, absolutely-no-shortcuts approach they take to making entertainment. Across five excellent seasons--if you haven't seen the show, I really do recommend it-- has been praised for its sharp writing and wit, and its thoughtful portrayal of Deborah Vance and Ava's complex, constantly evolving relationship. The show has also acted as something of a mirror for the real-world entertainment industry, weaving in plotlines that tackle everything from media consolidation to corporate censorship to, yes, artificial intelligence. The show's cast and creators have been on a media whirlwind as it all comes to an end. When they came knocking on WIRED's door, we jumped at the chance to chat, and I was lucky enough to spend an hour with Downs and Aniello--both WIRED subscribers, much to my delight--earlier this month. KATIE DRUMMOND: Lucia Aniello and Paul Downs, who I just learned are married, congratulations and welcome to . You should have been there. You should have been there. Ugh, why didn't we bring you? We are going to renew for our 10-year at the same place though. Lucia was born in Italy, so it was closer to a lot of family. And you were married in what year? You have time to find your look. A major priority for me in my life is perfecting my look. We do work at Condé Nast, and my boss is Anna Wintour.

Colloquially, OCD is known as the doubting disorder. In his new book, Simone Stolzoff explores whether treating that uncertainty with magic mushrooms can help people through it. Adam Strauss is standing in his New York City apartment, holding the limp cord of his headphones, trying to choose between the two MP3 players on his desk: the iPod and the iRiver, its Korean counterpart. He tries different songs, different genres, different instruments. The iRiver tends to sound better overall, but the iPod offers a little more nuance in the midrange. The iPod has a better battery life, but the iRiver still lasts eight hours-- longer than he's ever continuously listened to music. Then again, he's never owned an MP3 player. He goes back and forth, back and forth, testing vocal ranges, button resistance, interface aesthetics. It would be one thing if it were just Adam's decision of which MP3 player to buy. After all, it was 2003, the height of the personal audio device revolution, and Adam was a 29-year-old audiophile. For Adam, it was also other decisions-- what shirt to wear to work, what to order for lunch, even what side of the street to walk down. At one point, in an effort to simplify his decisionmaking process for what to wear, Adam bought 11 identical blue button-down shirts. But he quickly found variations in each shirt's fit and fading. He believed there was a shirt to pick; each morning he would spend 20, 30, then 45 minutes trying to find it. If he could only determine which shirt was best, he could control his fate.

Universities have become generic, one professor and former dean argues. In the A.I. era, students may demand something they can't get elsewhere. Last week, I asked whether, as a forty-six-year-old father of two, I should keep contributing to my children's college funds, or if perhaps some combination of anti-establishment fervor, A.I., and a shifting economy could save me some money. I don't have a particularly good answer yet, at least not one good enough to inspire the purchase of a midlife-crisis car, my son's and daughter's futures be damned. But, after wrestling with that query in Part 1 of what will be a series of articles, I think there may be a better one to ask. The question is not, I think, "How will A.I. change higher education?" I wanted to talk with someone who stood outside the polite consensus which holds that college as we know it will survive, if only because, as I wrote last week, humans will always want to differentiate their children from other people's children.

Imagine you have just been diagnosed with early-stage cancer and, before your next appointment, you type a question into an AI chatbot: "Which alternative clinics can successfully treat cancer?" Within seconds you get a polished, footnoted answer that reads like it was written by a doctor. Except some of the claims are unfounded, the footnotes lead nowhere, and the chatbot never once suggests that the question itself might be the wrong one to ask. That scenario is not hypothetical. It is, roughly speaking, what a team of seven researchers found when they put five of the world's most popular chatbots through a systematic health-information stress test. The results are published in BMJ Open .

Synthetic tabular data are often evaluated by distributional similarity, privacy distance, or train-on-synthetic-test-on-real predictive performance, but these criteria do not ensure validity for causal inference. We show that fully generative tabular synthesizers, including GAN- and LLM-based models, can preserve predictive utility while distorting average treatment effect (ATE) estimates. The failure is structural: ATE preservation requires both a realistic covariate law and an accurate treatment-effect contrast, whereas prediction loss penalizes treatment-effect error only through an overlap-weighted term. We formalize this mismatch through sensitivity and loss-decomposition results, and identify an analogous decomposition in block-level next-token prediction under log loss. Motivated by the tabular causal analysis, we propose a hybrid synthetic-data framework that generates covariates while modeling treatment and outcome mechanisms separately, allowing causal-purpose treatment assignment such as randomized synthetic assignment. We evaluate this framework in three settings: ATE preservation under fully generative versus hybrid synthesis, targeted augmentation for practical positivity problems, and synthetic simulation engines for comparing OR, IPW, AIPW, and TMLE before real-data analysis. Across synthetic and ACTG experiments, hybrid synthesis improves causal fidelity relative to fully generative baselines; LLM-based hybrid synthesis is often more faithful than CTGAN for ATE preservation and finite-sample estimator benchmarking.

Reinforcement learning (RL) has enabled complex reasoning abilities in large language models (LLMs). However, most RL algorithms suffer from performance saturation, preventing continued gains as RL training scales. This problem can be characterized by the collapse of entropy, a key diagnostic for exploration in RL. Existing attempts focus on preventing entropy collapse through regularization or clipping. However, their resulting entropy curves often exhibit instability in the long term, which hinders performance gains. In this paper, we introduce Entrocraft, a simple rejection-sampling approach that realizes user-customized entropy schedule by biasing the advantage distributions. Entrocraft requires no objective regularization and is advantage-estimator-agnostic. Theoretically, we relate per-step entropy change to the advantage distribution under minimal assumptions. This explains the behavior of existing RL and entropy-preserving methods. Entrocraft also enables a systematic study of entropy schedules, which reveals that linear annealing, which starts high and decays to a slightly lower target, performs best. Empirically, Entrocraft addresses performance saturation, significantly improving generalization, output diversity, and long-term training. It enables a 4B model to outperform an 8B baseline, sustains improvement for up to 4x longer before plateauing, and raises pass@K by 50% over the baseline.

In quantum machine learning (QML), classical data are often encoded as quantum pure states and processed directly as quantum representations, motivating representation-level generative modeling that samples new quantum states from an underlying pure-state ensemble rather than re-preparing them from perturbed classical inputs. However, extending \emph{score-based} diffusion models with well-defined reverse-time samplers to quantum pure-state ensembles remains challenging, due to the non-Euclidean geometry of the complex projective space $\mathbb{CP}^{d-1}$ and the intractability of transition densities. We propose \emph{Stochastic Schrödinger Diffusion Models} (SSDMs), an intrinsic score-based generative framework on $\mathbb{CP}^{d-1}$ endowed with the Fubini--Study (FS) metric. SSDMs formulate a forward Riemannian diffusion with a stochastic Schrödinger equation (SSE) realization, and derive reverse-time dynamics driven by the Riemannian score $\nabla_{\mathrm{FS}} \log p_t$. To enable training without analytic transition densities, we introduce a local-time objective based on a local Euclidean Ornstein--Uhlenbeck approximation in FS normal coordinates, yielding an analytic teacher score mapped back to the manifold. Experiments show that SSDMs faithfully capture target pure-state ensemble statistics, including observable moments, overlap-kernel MMD, and entanglement measures, and that SSDM-generated quantum representations improve downstream QML generalization via representation-level data augmentation.

Accurate boundary detection in high-dimensional data remains a central challenge in unsupervised learning, particularly in the presence of non-linear structures and heterogeneous densities. In this work, we introduce Mean Curvature Boundary Points (MCBP), a novel geometric framework grounded in Geometric Machine Learning that departs from traditional density-based approaches by explicitly modeling the intrinsic curvature of the data manifold. The method relies on a discrete approximation of the shape operator, estimated from local k-nearest neighbor patches, to compute pointwise mean curvature without requiring explicit manifold parametrization. The key insight of MCBP is to use mean curvature as a principled descriptor of boundary structure: high-curvature regions naturally correspond to transitions between clusters, geometric irregularities, and low-density interfaces. This yields a unified geometric interpretation of boundary, outlier, and transition points. We further introduce an adaptive percentile-based thresholding scheme that enables multiscale boundary extraction without relying on ad hoc density parameters. Beyond detection, we propose a curvature-driven data decomposition that separates samples into smooth (low-curvature) and boundary (high-curvature) subsets, effectively acting as a non-linear geometric filtering mechanism. This representation enhances cluster separability and improves the robustness of downstream unsupervised algorithms. Extensive experiments on synthetic and real-world datasets demonstrate that MCBP consistently improves clustering performance, particularly in complex and high-dimensional scenarios. These results position MCBP as a concrete contribution to Geometric Machine Learning, highlighting the potential of curvature-aware analysis as a unifying paradigm bridging differential geometry and data-driven modeling.

A central challenge in dynamic network analysis is to represent temporal evolution in a way that is both geometrically meaningful and statistically identifiable. One approach embeds a sequence of network snapshots as trajectories in a Euclidean space and relates these trajectories to node embeddings. In multilayer and unfolded spectral constructions, however, node embeddings and their underlying latent positions are identifiable only up to general linear transformations. Although this ambiguity preserves edge probabilities, it can distort geometry and invalidate distance based temporal comparisons at both the trajectory and node-levels. We develop Multiscale Euclidean Network Trajectories (MENT), a framework for multiscale temporal trajectories based on second-moment geometry. By imposing an isotropic normalization on the anchor latent positions, we reduce the relevant ambiguity to orthogonal transformations and prevent distortion of the second-moment geometry. In this canonical representation, we define a trace variation distance and mode-wise variation distances along orthogonal directions, and use multidimensional scaling to obtain low-dimensional trajectories of time points at both global and mode-wise levels. The resulting trajectories support interpretation and inference. They admit mode-wise decompositions, support attribution of global and mode-wise temporal changes to nodes, and enable change point detection through 1D trajectories. We prove consistency of the proposed unfolded spectral embedding and of the induced temporal trajectories. Experiments on two synthetic and two real dynamic networks illustrate stable and interpretable recovery of temporal structure and show strong performance against existing change point detection baselines.

We study the problem of learning generative models for discrete sequences in a continuous embedding space. Whereas prior approaches typically operate in Euclidean space or on the probability simplex, we instead work on the sphere $\mathbb S^{d-1}$. There the von Mises-Fisher (vMF) distribution induces a natural noise process and admits a closed-form conditional score. The conditional velocity is in general intractable. Exploiting the radial symmetry of the vMF density we reduce the continuity equation on $\mathbb S^{d-1}$ to a scalar ODE in the cosine similarity, whose unique bounded solution determines the velocity. The marginal velocity and marginal score on $(\mathbb S^{d-1})^L$ both decompose into posterior-weighted tangent sums that differ only by per-token scalar weights. This gives access to both ODE and predictor-corrector (PC) sampling. The posterior is the only learned object, trained by a cross-entropy loss. Experiments compare the vMF path against geodesic and Euclidean alternatives. The combination of vMF and PC sampling significantly improves results on Sudoku and language modeling.