Predicting the effect of interventions with many possible variations, e.g., therapeutic content that affects mental health outcomes or an earnings call transcript that drives movement in share price, is useful across several domains. However, classical causal estimators tend to assume that all possible interventions are observed, which is infeasible when interventions vary widely, for instance, in the space of all text strings. We adapt a well-known approach of recasting causal inference as a learning problem, to address high-dimensional treatment spaces. Specifically, under standard assumptions like no unobserved confounding, we show that causal error decomposes into a series of moment-balancing errors of increasing order, and design objectives that directly improve causal estimation. We also show how to project the effect of a high-dimensional treatment onto lower-dimensional treatment attributes, which allows a single model to answer several causal questions without additional attribute-specific training. We empirically evaluate our estimators in settings with high-dimensional continuous, discrete, and text treatments, the last of which used a semi-synthetic dataset of Amazon Reviews. Our experiments demonstrate the benefit of higher-order balance error optimization and competitive performance of projected causal estimates with attribute-specific estimators.

Reinforcement learning with verifiable rewards has become a standard recipe for improving the reasoning abilities of large language models. Existing algorithms face a tradeoff between computational efficiency and sample efficiency in value estimation and policy learning. We introduce BASIS, a critic-free post-training algorithm designed to address this tradeoff. At each online training step, BASIS samples only one rollout per prompt, but leverages rich information across prompts in the entire batch to improve value function estimation. Our experiments demonstrate that BASIS reduces MSE in value function estimation by 69% compared to REINFORCE++, a representative single-rollout baseline, and achieves lower MSE with one rollout than group mean estimators with 8 rollouts. This improvement in value estimation translates to better policy optimization: using substantially less training time, BASIS achieves performance close to multi-rollout GRPO-type baselines and often outperforms single-rollout REINFORCE-type baselines.

The Holy Father referenced in his encyclical about AI--an expert (if unintentional) troll of tech billionaires who keep misinterpreting the series. Nobody was surprised that Pope Leo XIV cited well-known saints and previous pontiffs in his first encyclical, or papal letter of spiritual guidance,, released Monday. But the name that immediately jumped out to many readers is one synonymous with high fantasy literature: J.R.R. Tolkien, the Catholic author of . Leo's letter is concerned with "safeguarding the human person in the time of artificial intelligence," a major theme of his first year as leader of the Catholic Church. Drawing from his predecessor, Pope Francis, he warns of "the growing dominance of a technocratic paradigm," one capable of "reducing creation to an object of exploitation and human beings to mere cogs in a system driven toward ever greater efficiency."

When Pope Leo XIV presented his first encyclical on artificial intelligence at the Vatican on Monday, he invited Christopher Olah, cofounder of Anthropic, to speak. The move signaled an unprecedented alliance between the Catholic church and Silicon Valley. But to understand how this partnership came about, we need to go back to Anthropic's founding. Anthropic launched in 2021 after a group of OpenAI researchers, including Dario and Daniela Amodei, left to form a rival lab. They did so with a clear conviction: Artificial intelligence models were becoming too powerful to be developed exclusively according to the logic of competition and speed.

Sam Altman Says AI'Jobs Apocalypse' He Once Predicted Probably Won't Happen. OpenAI CEO Sam Altman speaks during the BlackRock Infrastructure Summit on March 11, 2026 in Washington, DC. OpenAI CEO Sam Altman speaks during the BlackRock Infrastructure Summit on March 11, 2026 in Washington, DC. Throughout his rise to becoming one of the most influential CEOs in artificial intelligence, OpenAI's Sam Altman made repeated bold assertions about the impact that the new technology would have on jobs. He has said that AI will "probably replace most of the jobs people do today," that entire job categories will be "totally, totally gone," and that those impacted by the dramatic shifts will "find all sorts of new things to do. Now, however, Altman appears to have changed his tune, saying he is "delighted to be wrong" about the impact AI would have on employment. I don't think we're going to have the kind of jobs apocalypse that some of the companies in our space advocate or talk about, he said during a virtual interview at a Commonwealth Bank of Australia (CBA) conference in Sydney on Tuesday. "I thought there would have been more impact on entry-level white-collar jobs being eliminated by now than has actually happened, Altman said.

PCWorld examines Pope Leo XIV's encyclical on AI, which emphasizes that artificial intelligence reflects creator biases and lacks genuine empathy or real-world experience. The Pope calls AI a "valuable tool that requires vigilance," advising users to adopt a more thoughtful, slower-paced approach when interacting with models like ChatGPT, Claude, and Gemini. This papal guidance encourages users to actively consider when, why, and what they ask AI systems, recognizing their limitations despite sophisticated responses. Delving into Pope Leo XIV's exhaustive treatise about humanity and AI, I was struck by a recurring theme: AI simulates fundamental human traits that it doesn't actually possess. For starters, AI lacks the grounding we humans get from our real-world experiences, Pope Leo noted in his first encyclical, which was released Monday by the Vatican. Yes, AI models like ChatGPT (or more specifically, GPT), Claude, and Gemini are trained on mountains of data that seemingly represent the entirety of human knowledge. But all that data is just that: data.

Here's Exactly How That Happened The definitive story of how Claude Code and OpenClaw kicked off computing's biggest transformation possibly ever. "Hi, my name is Peter, and I'm a Claudeholic." It was August 2025 and Peter Steinberger was addressing a meetup in London called Claude Code Anonymous. Steinberger and some fellow addicts had arranged the event to network with people like themselves--techies swept up by coding tools such as Anthropic's paradigm-busting Claude Code. "I dedicate pretty much all my waking time to this, yet it doesn't feel enough," he told the gathering in a cozy, brick-walled room. A few months later, Anthropic released a new version of Claude Code, and the ranks of Claudeholics exploded . Called Opus 4.5, it could handle more complicated programming tasks, retain much more in its memory, run for many hours on end, and manage a team of AI subagents. Anthropic has what it describes as a "notoriously difficult" take-home exam for prospective engineering hires; in a head-to-head comparison of those people and its models, Anthropic claimed that Opus 4.5 "scored higher than any human candidate ever," which "raises questions on how AI will change engineering as a profession."

Offline Reinforcement Learning from Human Feedback (RLHF) pipelines such as Direct Preference Optimization (DPO) train on a pre-collected preference dataset, which makes them vulnerable to preference poisoning attack. We study label flip attacks against log-linear DPO. We first illustrate that flipping one preference label induces a parameter-independent shift in the DPO gradient. Using this key property, we can then convert the targeted poisoning problem into a structured binary sparse approximation problem. To solve this problem, we develop two attack methods: Binary-Aware Lattice Attack (BAL-A) and Binary Matching Pursuit Attack (BMP-A). BAL-A embeds the binary flip selection problem into a binary-aware lattice and applies Lenstra-Lenstra-Lovász reduction and Babai's nearest plane algorithm; we provide sufficient conditions that enforce binary coefficients and recover the minimum-flip objective. BMP-A adapts binary matching pursuit to our non-normalized gradient dictionary and yields coherence-based recovery guarantees and robustness (impossibility) certificates for $K$-flip budgets. Experiments on synthetic dictionaries and the Stanford Human Preferences dataset validate the theory and highlight how dictionary geometry governs attack success.

This paper introduces a physics-informed generative framework that resolves the fundamental conflict between the statistical flexibility of deep learning and the rigorous theoretical constraints of fixed-income modeling. We demonstrate that standard generative models and unconstrained statistical extrapolations suffer from "manifold collapse" and severe arbitrage violations when forecasting term structures across diverse macroeconomic regimes. To overcome this, we propose a two-stage architecture. First, a Student-t Conditional Variational Autoencoder with Dynamic Level Injection (CVAEsT+LS) extracts a robust, heavy-tailed term structure manifold, effectively decoupling macroeconomic shape dynamics from absolute base rates. Second, the latent dynamic evolution is governed by a continuous-time Neural Stochastic Differential Equation (SDE) strictly penalized by a No-Arbitrage Partial Differential Equation (PDE). Empirical results across multiple sovereign currencies (USD, GBP, JPY) confirm that our synergistic approach drastically reduces out-of-sample forecasting errors -- achieving an exceptional 6.58 bps Mean Tenor RMSE -- and successfully overcomes the massive parallel drift and zero-lower-bound violations exhibited by the classical HJM model in extreme environments. Furthermore, through phase space vector field analysis, we demonstrate the model's superior capability in unsupervised macroeconomic regime detection and high-quality continuous-time scenario generation. Ultimately, this research provides a highly scalable, mathematically sound evolutionary engine for term structure modeling.

Human feedback often arrives as preferences rather than calibrated numeric rewards, motivating reinforcement learning from preferential feedback, also referred to as reinforcement learning from human feedback (RLHF). We present a rigorous theoretical study of preference-only learning in episodic kernel MDPs. In each episode, the learner deploys two policies from a common start state and receives a single binary label indicating which trajectory is preferred, modeled by a Bradley--Terry--Luce link on the difference of cumulative (unobserved) rewards. Under kernel-based assumptions on the reward and transition functions (one of the most general models amenable to theoretical analysis) we develop preference-based value estimation and confidence sets tailored to end-of-episode comparisons. We prove high-probability regret bounds that scale sublinearly in the number of episodes, implying that the value of the learned policy converges to that of the optimal policy.