Amazon has pulled a video recap made with artificial intelligence (AI) from its hit TV show Fallout after users said it got several facts wrong about the series. The firm said in November it was testing the first-of-its-kind tool in the US to help viewers catch up on some of its shows on streaming service Prime Video - including Fallout, its adaptation of the popular video game franchise. But it has since disappeared from the site after users highlighted mistakes in its video summarising the events of Fallout season one - including claiming one scene was set more than 100 years earlier than it was. The BBC has approached Amazon for comment. The move to apparently press pause on its AI-powered recaps was first reported by tech publication The Verge .

Amazon's AI-generated recap for the Fallout TV series contained a major factual error, incorrectly stating the Great War occurred in the 1950s instead of 2077. PCWorld reports that Amazon removed the erroneous recap after Games Radar spotted the mistake, along with similar AI-generated content for other shows. This incident highlights risks of using AI without proper oversight, potentially damaging brand credibility through easily avoidable factual mistakes.

The company launched the video recap feature in November, and it's already getting the details wrong. Amazon's plan to offer AI-generated recaps of Prime Video shows isn't off to a great start. The company's recap of the first season of features multiple errors, writes, including basic facts about the plot of the show. You can watch the recap yourself in the Extras section of Amazon's season two listing in Prime Video. Besides being somewhat dry, the AI-generated recap incorrectly identifies the time period of the show's Los Angeles-set flashbacks as being the 1950s, when they're actually 2077 (the Fallout franchise is set in an alternate history that diverged from our real one after 1945). As notes, the recap also seems to misunderstand the ending of the first season, which sets up season two's partnership between vault dweller Lucy and The Ghoul, an irradiated wastelander with a personal connection to the mystery at the heart of the first season.

Large language models in healthcare often miss critical emotional cues, delivering medically sound but emotionally flat advice. Such responses are insufficient in clinical encounters, where distressed or vulnerable patients rely on empathic communication to support safety, adherence, and trust. We present RECAP (Reflect-Extract-Calibrate-Align-Produce), an inference-time framework that guides models through structured emotional reasoning without retraining. RECAP decomposes patient input into appraisal-theoretic stages, identifies psychological factors, and assigns Likert-based emotion likelihoods that clinicians can inspect or override, producing nuanced and auditable responses. Across EmoBench, SECEU, and EQ-Bench, RECAP improves emotional reasoning by 22-28% on 8B models and 10-13% on larger models over zero-shot baselines. In blinded evaluations, oncology clinicians rated RECAP's responses as more empathetic, supportive, and context-appropriate than prompting baselines. These findings demonstrate that modular, principled prompting can enhance emotional intelligence in medical AI while maintaining transparency and accountability for clinical deployment.

"intuition behind the statistical dimension" and how it varies between kernels

If we cannot inspect the training data of a large language model (LLM), how can we ever know what it has seen? We believe the most compelling evidence arises when the model itself freely reproduces the target content. As such, we propose RECAP, an agentic pipeline designed to elicit and verify memorized training data from LLM outputs. At the heart of RECAP is a feedback-driven loop, where an initial extraction attempt is evaluated by a secondary language model, which compares the output against a reference passage and identifies discrepancies. These are then translated into minimal correction hints, which are fed back into the target model to guide subsequent generations. In addition, to address alignment-induced refusals, RECAP includes a jailbreaking module that detects and overcomes such barriers. We evaluate RECAP on EchoTrace, a new benchmark spanning over 30 full books, and the results show that RECAP leads to substantial gains over single-iteration approaches. For instance, with GPT-4.1, the average ROUGE-L score for the copyrighted text extraction improved from 0.38 to 0.47 - a nearly 24% increase.

Long-horizon tasks requiring multi-step reasoning and dynamic re-planning remain challenging for large language models (LLMs). Sequential prompting methods are prone to context drift, loss of goal information, and recurrent failure cycles, while hierarchical prompting methods often weaken cross-level continuity or incur substantial runtime overhead. We introduce ReCAP (Recursive Context-Aware Reasoning and Planning), a hierarchical framework with shared context for reasoning and planning in LLMs. ReCAP combines three key mechanisms: (i) plan-ahead decomposition, in which the model generates a full subtask list, executes the first item, and refines the remainder; (ii) structured re-injection of parent plans, maintaining consistent multi-level context during recursive return; and (iii) memory-efficient execution, bounding the active prompt so costs scale linearly with task depth. Together these mechanisms align high-level goals with low-level actions, reduce redundant prompting, and preserve coherent context updates across recursion. Experiments demonstrate that ReCAP substantially improves subgoal alignment and success rates on various long-horizon reasoning benchmarks, achieving a 32% gain on synchronous Robotouille and a 29% improvement on asynchronous Robotouille under the strict pass@1 protocol.

Large reasoning models (LRMs) "think" by generating structured chain-of-thought (CoT) before producing a final answer, yet they still lack the ability to reason critically about safety alignment and are easily biased when a flawed premise is injected into their thought process. We propose RECAP (Robust Safety Alignment via Counter-Aligned Prefilling), a principled reinforcement learning (RL) method for post-training that explicitly teaches models to override flawed reasoning trajectories and reroute to safe and helpful responses. RECAP trains on a mixture of synthetically generated counter-aligned CoT prefills and standard prompts, requires no additional training cost or modifications beyond vanilla reinforcement learning from human feedback (RLHF), and substantially improves safety and jailbreak robustness, reduces overrefusal, and preserves core reasoning capability -- all while maintaining inference token budget. Extensive analysis shows that RECAP-trained models engage in self-reflection more frequently and remain robust under adaptive attacks, preserving safety even after repeated attempts to override their reasoning.

Creators struggle to edit long-form, narrative-rich videos not because of UI complexity, but due to the cognitive demands of searching, storyboarding, and sequencing hours of footage. Existing transcript- or embedding-based methods fall short for creative workflows, as models struggle to track characters, infer motivations, and connect dispersed events. We present a prompt-driven, modular editing system that helps creators restructure multi-hour content through free-form prompts rather than timelines. At its core is a semantic indexing pipeline that builds a global narrative via temporal segmentation, guided memory compression, and cross-granularity fusion, producing interpretable traces of plot, dialogue, emotion, and context. Users receive cinematic edits while optionally refining transparent intermediate outputs. Evaluated on 400+ videos with expert ratings, QA, and preference studies, our system scales prompt-driven editing, preserves narrative coherence, and balances automation with creator control.

HBO's The Last of Us showed viewers in season one that it would lean heavily on the source video games for major plot points and general direction of the season while expanding on the universe, and season two has followed that to the most extreme end possible. Episode two sees Tommy and Maria lead the town of Jackson Hole against a massive wave of Infected, the likes of which we haven't seen in the show (or video games) yet. This was a complete invention for the show, one that gives the episode Game of Thrones vibes, or calls to mind a battle like the siege of Helm's Deep in Lord of the Rings: The Two Towers. It's epic in scale, with the overmatched defenders showing their skill and bravery against overwhelming odds; there is loss and pain but the good guys eventually triumph. That mass-scale battle is paired with the most intimate and brutal violence we've seen in the entire series so far, as Joel's actions finally catch up with him.