Amazon's Zoox recalls self-driving vehicles amid emergency response issues The Amazon subsidiary company Zoox has said that it will recall its fleet of 105 autonomous vehicles in the United States. The technology company announced the recall on Friday due to mounting concerns that the vehicles may not detect heavy smoke and could impede emergency personnel. Zoox said on Friday that on June 20 an unoccupied Zoox autonomous vehicle encountered heavy smoke that obscured an active emergency fire scene. The Zoox vehicle entered the scene, then braked hard while attempting to steer away, before coming to a stop. The Zoox vehicle, under teleguidance, reversed, after which first responders placed traffic cones at the scene, blocking two of the three lanes.

Follow this section to personalize your feed and get instant alerts. Follow Go to your personalized feed WHY FOLLOW? Smart Alerts: Get notified about major news as it happens. Follow this tag to personalize your feed and get instant alerts. Follow Go to your personalized feed WHY FOLLOW?

A central challenge in language models (LMs) is faithfulness hallucination: the generation of information unsubstantiated by input context. To study this problem, we propose Precise Information Control (PIC), a new task formulation that requires models to generate long-form outputs grounded in a provided set of short self-contained statements, without adding any unsupported ones. PIC includes a full setting that tests a model's ability to include exactly all input claims, and a partial setting that requires the model to selectively incorporate only relevant claims. We present PIC-Bench, a benchmark of eight long-form generation tasks (e.g., summarization, biography generation) adapted to the PIC setting, where LMs are supplied with well-formed, verifiable input claims. Our evaluation of a range of open and proprietary LMs on PIC-Bench reveals that, surprisingly, state-of-the-art LMs still hallucinate against user-provided input in over 70% of generations. To alleviate this lack of faithfulness, we introduce a post-training framework that uses a weakly supervised preference data construction method to train an 8BPIC-LM with stronger PIC ability--improving from 69.1% to 91.0% F1 in the full PIC setting. When integrated into end-to-end factual generation pipelines, PIC-LM improves exact match recall by 17.1% on ambiguous QA with retrieval, and factual precision by 30.5% on a birthplace fact-checking task, underscoring the potential of precisely grounded generation.

Computational models support high-stakes decisions across engineering and science, and practitioners increasingly seek probabilistic predictions to quantify uncertainty in such models. Existing approaches generate predictions either by sampling input parameter distributions or by augmenting deterministic outputs with uncertainty representations, including distribution-free and distributional methods. However, sampling-based methods are often computationally prohibitive for real-time applications, and many existing uncertainty representations either ignore input dependence or rely on restrictive Gaussian assumptions that fail to capture asymmetry and heavy-tailed behavior. Therefore, we extend the ACCurate and Reliable Uncertainty Estimate (ACCRUE) framework to learn input-dependent, non-Gaussian uncertainty distributions, specifically two-piece Gaussian and asymmetric Laplace forms, using a neural network trained with a loss function that balances predictive accuracy and reliability. Through synthetic and real-world experiments, we show that the proposed approach captures an input-dependent uncertainty structure and improves probabilistic forecasts relative to existing methods, while maintaining flexibility to model skewed and non-Gaussian errors.

The novelist Ewan Morrison was alarmed, though amused, to discover he had written a book called Nine Inches Pleases a Lady. Intrigued by the limits of generative artificial intelligence (AI), he had asked ChatGPT to give him the names of the 12 novels he had written. "I've only written nine," he says. "Always eager to please, it decided to invent three." The "nine inches" from the fake title it hallucinated was stolen from a filthy Robert Burns poem.

Morrisons has unveiled its newest staff members - in the form of aisle-roaming robots. The retail giant is trialling'Tally' robots at three stores in Wetherby, Redcar and Stockton, to monitor how products are being displayed on shelves. Using advanced AI and computer vision technology, Tally is designed to spot out-of-stock items, pricing errors, and misplaced products. Morrisons' technology manager, Katherine Allanach, called this a'crucial' role. 'It is a crucial but time-consuming task and so Tally aims to allow more time for colleagues to focus on customer service,' she told The Grocer.

Buying something in the shops used to be as simple as choosing the item and handing over the money. But in recent years, the great British shopping experience has dramatically changed. In 2025, artificial intelligence (AI) is patrolling Britain's retail stores to keep an eye on customers as they stock up on essentials. Now, people are subjected to a slew of AI-powered tech, including intelligent surveillance cameras, robots, facial recognition systems and online age checks. Home Bargains is the latest to follow the trend, with a new AI-enabled security system that watches you while you scan your own items.

The 2025 Oscar nominations were announced on Thursday, and two New Yorker films are among the contenders. "Incident," which uses body-camera and surveillance footage to examine a police shooting in Chicago, is nominated in the Documentary Short Film category, while "I'm Not a Robot," a darkly humorous Dutch film about a woman taking a series of CAPTCHA tests, is nominated for best Live Action Short. Seventeen previous New Yorker films have been nominated for Academy Awards; a victory at this year's ceremony, scheduled for March 2nd in Los Angeles, would be the magazine's first win. "Incident," directed by Bill Morrison, who produced with Jamie Kalven, chronicles a police killing and its aftermath. On a Chicago sidewalk, an African American man named Harith (Snoop) Augustus is questioned and then pursued by a foot patrol after leaving the barbershop where he works; after a brief scuffle, he is fatally wounded.

A Scottish man felt like he was being'kidnapped' when his electric vehicle (EV) appeared to develop a mind of its own at the weekend. Brian Morrison, 53, was heading home from work when his £30,000 MG ZS suffered a'catastrophic malfunction' on the A803 towards Kirkintilloch, near Glasgow. He was left terrified when the brake pedals stopped working and the car – which is designed and built in China – began driving itself at 30mph. Because he has mobility issues, Mr Morrison was unable to jump out of the car, which only came to a stop after it was deliberately driven into a police van. Thankfully no-one was hurt - but the incident has raised questions of whether the same issue could affect other EV users.

We introduce a frustratingly simple, super efficient and surprisingly effective decoding method, which we call Frustratingly Simple Decoding (FSD), for neural text generation. The idea behind FSD is straightforward: we build an anti-LM based on previously generated text and use this anti-LM to penalize future generation of what has been generated. The anti-LM can be implemented as simple as an n-gram language model or a vectorized variant. In this way, FSD introduces no extra model parameters and negligible computational overhead (FSD can be as fast as greedy search). Despite the simplicity, FSD is surprisingly effective; Experiments show that FSD can outperform the canonical methods to date (i.e., nucleus sampling) as well as several strong baselines that were proposed recently.