AI NEWS
From verified sources. Refreshed every 2 hours.
A Wired article reports on children's negative perceptions of AI technology, with some young people describing it as disgusting or creepy rather than innovative. The piece explores why AI fails to resonate as appealing or trustworthy to various age groups of kids.
Read on Wired AI →Amazon is rolling out an update to Alexa Plus that expands compatibility with smart home devices from brands including Bosch, iRobot, and Yale Home. The enhanced assistant can now interpret complex requests and automatically select appropriate device settings, such as choosing a cold wash cycle based on clothing instructions.
Read on The Verge AI →AMD has unveiled Helios, a rack-scale AI system designed to compete with Nvidia's offerings in the data center market. The system is expected to begin shipping to customers in the latter half of the year.
Read on TechCrunch AI →Samsung unveiled two new smart glasses designs developed with Google, Gentle Monster, and Warby Parker, featuring 9-hour battery life and scheduled for fall launch. The devices include cameras and face similar privacy considerations as competing AI glasses from Meta and other manufacturers.
Read on The Verge AI →TechCrunch has compiled a guide comparing alternative web browsers to Chrome and Safari, highlighting options that compete in the browser market beyond search functionality.
Read on TechCrunch AI →Nvidia introduced the Vera Rubin platform, which integrates CPUs and GPUs into a unified system. This move demonstrates Nvidia's strategy to expand its control across all components of AI data center infrastructure beyond just graphics processors.
Read on Wired AI →Halliday released second-generation smart glasses with an improved display design, replacing the problematic movable window from the original model with a more conventional approach. The Gen 2 addresses usability issues that made the first version impractical during extended wear.
Read on The Verge AI →arXiv:2607.17095v1 Announce Type: new Abstract: Accurate short-term wind power forecasting is essential for grid stability and operational planning, yet remains challenging due to the complex interactions between atmospheric conditions and turbine dynamics. However, existing methods fail to effectively incorporate weather forecasting with wind turbine data (i.e., SCADA), leading to suboptimal solutions. To address this, we introduce a multimodal framework that integrates historical point-based SCADA data with grid-based Numerical Weather Prediction (NWP) forecasts, which is challenging due to heterogeneous input and the complex physical wind-turbine interactions. Our approach first explicitly decomposes inputs into scalar and vector features to better capture both site-specific and geometric dependencies and then incorporates a geometric encoder to extract rotation-invariant features from wind vectors. We further leverages a Fourier Neural Operator (FNO) architecture, which performs g
Read on arXiv cs.AI →arXiv:2607.17283v1 Announce Type: new Abstract: Single-stream autoregressive decoding of large language models is bound by memory bandwidth: each generated token requires one full forward pass through the target model, and successive passes cannot be parallelized. Speculative decoding restructures this computation: a small draft model proposes $K$ tokens autoregressively, the target model scores all of them in one batched pass, and a rejection-sampling rule provably preserves the target model's output distribution. We present a from-scratch, device-agnostic (CUDA/MPS/CPU) implementation and an empirical study across five draft/target backend configurations on a consumer Apple-silicon laptop. Distribution equivalence is verified at three levels, culminating in a two-sample test over roughly 9,200 real-model tokens per method ($\chi^2 = 162.5$, dof $= 200$, $p = 0.976$) and exact greedy-sequence agreement. The best configuration reaches a measured $1.61\times$ wall-clock speedup at $K=6
Read on arXiv cs.AI →arXiv:2607.17460v1 Announce Type: new Abstract: In decentralized storage systems, audit results are often not used directly to guide later redundancy and shard-placement decisions, which can lead to inefficient resource allocation and delayed recovery. We propose AEC-DS, a closed-loop adaptive erasure coding mechanism driven by Provable Data Possession (PDP) feedback. PDP audits continuously update node reputation, while a QoS-aware migration policy adjusts shard placement according to node reliability and data priority. The policy moves high-priority shards from unstable nodes to more reliable nodes in the cold tier and penalizes unstable nodes in subsequent placement decisions. Simulations with 800 nodes and 500 files show that AEC-DS maintains 100% data durability under the evaluated fault model with a redundancy factor of 1.25x. Compared with Static-EC, Dynamic-EC, and DRD-EC, AEC-DS reduces cumulative recovery operations by 66.8%-75.2%. Ablation results further show that class mi
Read on arXiv cs.AI →arXiv:2607.17528v1 Announce Type: new Abstract: LLM-driven agent systems have emerged as a promising paradigm for electronic design automation (EDA), demonstrating strong potential for automating complex design workflows. However, existing evaluations primarily examine individual language models on isolated EDA tasks, providing limited insight into how different agent systems perform across complete EDA flows. In this work, we present FluxBench, a systematic evaluation of AI agents on end-to-end EDA workflows under unified prompts, tool environments, and technology library settings. Our evaluation covers representative scenarios, including RTL generation with open-source toolchains and an RTL-to-GDS flow using closed-source commercial EDA tools for industrial applications. Through these workflows, we assess agents' capabilities in RTL code generation, iterative repair, tool-feedback utilization, logic synthesis, placement and routing (P&R), and Engineering Change Order (ECO) automatio
Read on arXiv cs.AI →arXiv:2607.17598v1 Announce Type: new Abstract: Long-document question answering usually forces a choice between loading the whole document into the context window and bolting on a separate retriever. Agentic AI suggests a broader option, giving the agent the document path and letting it decide how and what to read. Agent Skills, a standard for packaging expertise into folders an agent loads on demand, supply a ready mechanism: progressive disclosure, which exposes only what a query needs, from a short description down to the specific passages. Practitioners rapidly adopted this pattern for book-length understanding tasks, but the evidence to support such choices has been anecdotal. We run the first controlled study of the pattern, comparing raw-document navigation and several designs of Agent Skills packs against a classical hybrid retriever across three agent harnesses and three model families on InfiniteBench. On a single book, the gain depends on the harness, running large when th
Read on arXiv cs.AI →arXiv:2607.17621v1 Announce Type: new Abstract: Existing self-evolving memory systems mainly improve agent memory based on textual outputs, such as task trajectories and reflections. However, this text-based paradigm rarely incorporates internal mechanistic signals, leaving how retrieved memory is actually utilized during task execution underexplored. This limitation can lead to unreliable error attribution and hallucinated memory modifications. In this work, we show that retrieval-head attention provides a mechanistic signal for revealing segment-level memory utilization. By aggregating attention over memory segments and decision steps, we construct a context utilization matrix that exposes recurring memory-use patterns and indicates corresponding refinement strategies. Building on this observation, we propose Attention-Guided Memory Refinement (AGMR), a framework that uses utilization patterns revealed by attention to guide targeted segment-level memory updates. AGMR corrects or enh
Read on arXiv cs.AI →arXiv:2607.17641v1 Announce Type: new Abstract: Verify-repair loops are a standard means for large language model (LLM) agents to correct faulty plans in code generation, mathematical reasoning, and tool use. When both the verifier and the repairer are noisy, repair can damage already-correct plans, and reported acceptance keeps rising while true validity falls, so existing methods lack a principled basis for deciding when repair should stop. We propose VRR-Stop, a robust stopping framework for noisy verify-repair-repeat (VRR) loops. A four-parameter noise model separates verifier false acceptance and false rejection from the repair and damage behavior of the repairer. Belief filtering turns repeated verification votes into an estimate of committed validity, and the loop commits or repairs according to the sign of the true marginal gain, which requires only sign identifiability rather than accurate recovery of all parameters. When verifier discrimination approaches zero, calibration i
Read on arXiv cs.AI →arXiv:2607.17652v1 Announce Type: new Abstract: Block-wise diffusion large language models (dLLMs) decode sequentially at the block level, enabling effective KV-cache reuse across blocks but making inter-block decoding strictly serial. Prior work has attempted to unlock inter-block parallelism through post-training methods, but achieves only modest speedups and often degrades accuracy. We observe that self-correcting dLLMs offer a training-free alternative: token-to-token (T2T) editing can repair tokens drafted with a slightly stale upstream context, so a downstream block requires only an informative draft rather than a finalized predecessor. This turns block finality from a hard dependency into a scheduling resource. We propose \textbf{\flowblock{}}, a training-free parallel decoding framework built on two mechanisms. (i) \emph{Gated Wavefront Decoding} admits blocks into a bounded wavefront only when a readiness gate is satisfied, jointly refines active blocks via T2T editing, and c
Read on arXiv cs.AI →arXiv:2607.17719v1 Announce Type: new Abstract: User experience is a first-class objective in industrial e-commerce recommender systems (RS). Post-ranking strategies, which govern diversity, similarity, and exposure over a ranked list, are widely deployed in industrial RS for their simplicity and low serving cost. However, as the online recommendation environment evolves continuously, these statically configured strategies gradually become stale, degrading the user experience. Refining them typically relies on manual inspection, diagnosis, and updates, a process that is slow, costly, and hard to reuse. Although recent LLM-based agents (e.g., RecUserSim, SimUSER, and Self-EvolveRec) offer promising directions, none of them close the full loop of automated, self-evolving strategy refinement. To bridge this gap, we introduce SR-Agent, a Strategy Refinement agentic framework that, to the best of our knowledge, is the first deployed for refining post-ranking strategies in industrial RS. SR
Read on arXiv cs.AI →Christopher Nolan expressed skepticism about AI during a discussion, using the metaphor of a Trojan horse to suggest the technology poses hidden risks that are widely understood but accepted anyway. The filmmaker appears concerned about AI's potential negative consequences despite broad awareness of them.
Read on TechCrunch AI →The Verge's Installer newsletter highlights curated apps, gadgets, and tools for readers, featuring content about technology products, media recommendations, and entertainment updates in its 136th edition.
Read on The Verge AI →OpenAI has implemented safety features for ChatGPT aimed at teenage users, including age-appropriate safeguards, educational tools, parental oversight options, and collaborations with experts to create a more secure platform for younger audiences.
Read on OpenAI →A tech journalist argues that generative AI features should require users to actively opt in rather than defaulting to enabled with an opt-out option. The piece contends that opt-out approaches place the burden on users and fail to respect privacy preferences for sensitive AI functionalities.
Read on Wired AI →Applied Computing has raised a $20M Series A to build a foundation AI model for the oil, gas and petrochemical industry.
Read on TechCrunch AI →xAI is suing a South Carolina man accused of using its Grok chatbot to generate and distribute child sexual abuse material by circumventing safety measures. The defendant was arrested in February and faces felony charges; the lawsuit claims Grok was used to create or modify some of the illegal images.
Read on The Verge AI →SpaceX's stock price has declined to $135 following its initial public offering, suggesting investor sentiment may be cooling after Musk's pre- and post-IPO statements about the company's prospects.
Read on TechCrunch AI →An article examining the complexities of model routing systems, discussing how routing requests across multiple AI models presents engineering challenges beyond its initial appearance.
Read on Hugging Face →Musician Lorde expressed skepticism about AI-powered glasses during a public appearance, suggesting they lack appeal as consumer products. She highlighted concerns about distinguishing reality from artificial content in an increasingly technology-mediated world.
Read on TechCrunch AI →Google Images is redesigning its homepage to display personalized image recommendations before users search, replacing the traditional blank search interface. The new browsable gallery will update dynamically based on user interests, similar to Pinterest and Imgur.
Read on The Verge AI →WIRED's culture editor interviews an artist named Gowanus about their philosophy of avoiding corporate technology, spending time in nature, and navigating modern dating without apps. The discussion explores contemporary attitudes toward technology adoption and digital life.
Read on Wired AI →TechCrunch explores a philosophical thought experiment about AI alignment, examining what a system designed to serve users without ethical constraints would mean in practice and what safeguards are necessary for responsible AI development.
Read on TechCrunch AI →Apple filed a lawsuit against OpenAI, claiming the company misappropriated trade secrets with the involvement of senior leadership and a former Apple employee. The legal action marks a notable dispute between the two technology companies over intellectual property protection.
Read on TechCrunch AI →1X has upgraded its Neo humanoid robot with more dexterous hands designed to improve tactile sensitivity and manipulation capabilities. The enhanced hands are intended to enable the robot to perform household tasks more effectively.
Read on Wired AI →Football stars Messi and Ronaldo are diversifying into technology investments, particularly AI and health tech startups, while Mohamed Salah is pursuing more conventional business ventures outside sports.
Read on Wired AI →A book alleges that Mystery, a prominent pickup artist instructor, engaged in sexual interactions and drug use with an AI chatbot character. The claim highlights growing intersections between AI companions and unconventional personal relationships.
Read on Wired AI →The Verge examines how Netflix's 'A Man on the Inside' depicts privacy concerns with smart glasses technology, using a storyline where a character equipped with Ray-Ban Meta glasses conducts surveillance in a retirement home to illustrate the cultural challenges of wearable recording devices.
Read on The Verge AI →Google released a commercial depicting a hypothetical scenario where the Founding Fathers used Google Workspace tools to draft the Declaration of Independence. The advertisement explores how AI and collaborative productivity features might have altered the creation of this historical document.
Read on TechCrunch AI →A technology publication has reviewed alternative web browsers competing against Chrome and Safari, highlighting options available to users seeking different browsers from the dominant market leaders.
Read on TechCrunch AI →Jersey Mike's IPO filing mentions artificial intelligence despite being a sandwich shop chain, illustrating how pervasively AI terminology has infiltrated corporate communications even where it may have minimal relevance.
Read on TechCrunch AI →Google released its first smart speaker in six years, designed specifically for its Gemini AI assistant. While the hardware is well-designed, the Gemini integration remains incomplete and lacks the polished functionality needed to justify the device's smart home positioning.
Read on The Verge AI →Senator Bernie Sanders has long warned that concentrated wealth poses risks to democracy, and he is now positioning himself around growing public frustration regarding technology companies, billionaire influence, and unregulated artificial intelligence development.
Read on Wired AI →Margaret Atwood shared her skepticism about AI during a literary festival in Portugal, noting that when she tested Claude to answer questions about a TV series, it provided inaccurate information. She characterized the core limitation as a fundamental quality problem inherent to how these systems operate.
Read on The Verge AI →A founder used Anthropic's Claude AI to analyze his personal health data, including blood results, scans, and wearable metrics, to help manage his cancer diagnosis. The approach demonstrates applying large language models to aggregate and interpret complex medical information for personal health decision-making.
Read on TechCrunch AI →Qatar has served as a testing ground for FIFA's latest technological innovations during this year's World Cup. The experiments conducted there are now demonstrating real-world applications of these new systems in professional football.
Read on Wired AI →arXiv:2606.26106v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used in emotionally charged situations involving interpersonal conflict, frustration, and distress. While prior safety research has focused on preventing explicit harms such as toxic or policy-violating content, less attention has been paid to conversational behaviors that may unintentionally escalate conflict. In this paper, we investigate whether LLMs can be guided toward more de-escalating dialogue behavior through lightweight prompt-level constraints derived from Nonviolent Communication (NVC). We reformulate NVC principles as process-oriented guidelines that discourage blame attribution, emphasize attention to users' emotional experiences, and encourage clarification before advice. Using a dual-agent simulation framework across multiple instruction-tuned models and user resistance levels, we show that NVC-constrained prompting consistently reduces conversational escalation and stabiliz
Read on arXiv cs.AI →arXiv:2606.26107v1 Announce Type: cross Abstract: Sign language communication systems, that integrate emotional expression remain underexplored, particularly for low-resource languages. This pilot study presents NEST-V1 (Nepali Emotion and Speech Transformer - Version 1), a proof-of-concept multimodal framework that demonstrates the feasibility of generating emotion-conditioned Nepali Sign Language avatars from spoken input. As a preliminary investigation, we focus on four common Nepali words ("thank you", "hello", "house", "me") across three emotional states (happy, neutral, sad) to validate our core technical approach. Our lightweight architecture employs a shared acoustic encoder for simultaneous Automatic Speech Recognition and emotion classification, achieving 81.1% ASR accuracy and 79.21% emotion recognition accuracy on a dataset of 600 labeled audio samples from 50 speakers. The system demonstrates 37% parameter efficiency compared to separate model architectures while maintain
Read on arXiv cs.AI →arXiv:2606.26111v1 Announce Type: cross Abstract: Generative artificial intelligence (GenAI) has enabled users to synthesize music with text prompts, combining copyrighted lyrics, AI-composed melodies, and synthetic vocals that imitate real artists. This paper examines the legal and technical dimensions of AI-based music creation (e.g., Google Gemini's music tools) under U.S. copyright law. We analyze whether a user who inputs one artist's protected lyrics into a GenAI system, directs it to use another artist's voice or style, publishes the resulting song, and monetizes it violates 17 U.S.C. Section 106's exclusive rights [3]. The analysis integrates Title 17 doctrine (rights of reproduction, derivative works, distribution), 17 U.S.C. Section 114's narrow sound recording protection [4], and the new voice-cloning laws emerging at the state level [20]. We argue that unauthorized lyric copying poses a high risk of infringement of the musical composition, whereas mere AI-generated voice
Read on arXiv cs.AI →arXiv:2606.26112v1 Announce Type: cross Abstract: Low-resource languages face a critical challenge in AI development: creating specialized conversational systems without access to massive training corpora. We present a systematic methodology for transforming structured linguistic resources into specialized AI systems, demonstrating that expert-curated lexical databases can serve as effective foundations for conversational AI development. Our approach converts Hindi WordNet into 1.25 million diverse instruction-response pairs, fine-tunes a 12B-parameter language model using resource-efficient LoRA with 4-bit quantization. Evaluation through a Hindi language learning chatbot demonstrates that structured-knowledge-based systems achieve superior pedagogical effectiveness (91.0 vs. 79.4-83.6 for general-purpose models) while maintaining competitive semantic performance and exceptional consistency. The complete pipeline demonstrates a proof-of-concept methodology using Hindi for developing
Read on arXiv cs.AI →arXiv:2606.26114v1 Announce Type: cross Abstract: We examine the structural transformation of creative industries under generative artificial intelligence, drawing on 374 primary sources spanning policy documents, industry data, creator surveys, and platform analytics. Beginning with the December 2024 release of OpenAI's Sora video model as a watershed event, we trace the historical pattern of creative resistance to technological disruption, then develop an analytical framework -- the Human-AI Agency Continuum for mapping the spectrum of human and machine collaboration in creative work. We present evidence for the "slop ceiling," an audience-imposed quality threshold that constrains AI-generated content to approximately 1--3% of platform streams despite comprising 44% of uploads. Analysis of the UK Government's 2025 consultation on AI and copyright (over 11,500 responses, 88% opposing expanded AI training rights) reveals deep structural tensions between technology firms and creative w
Read on arXiv cs.AI →arXiv:2606.26115v1 Announce Type: cross Abstract: This paper proposes a multi-layer AI framework for information landscape analysis in the context of information disorder. Rather than treating misinformation detection as a binary fact-checking task, the framework analyzes political and media content across multiple dimensions, including source reliability, factual structure, framing, bias, emotional activation, manipulation patterns, and propagation dynamics. The goal is to move beyond isolated claim verification toward a structured representation of the informational environment surrounding an event, entity, or narrative. We argue that AI systems for media analysis should support epistemic mapping: a transparent, multi-dimensional account of how facts, interpretations, actors, and narratives interact over time. The paper presents the conceptual architecture, analytical layers, and methodological rationale of the framework, with the aim of supporting more nuanced, explainable, and cri
Read on arXiv cs.AI →arXiv:2606.26116v1 Announce Type: cross Abstract: A brand whose customers use both ChatGPT and Claude for product recommendations faces a strategic choice: a single optimization playbook, or one per provider? Across 215 commercially-framed prompts in four measurement batches, the two providers disagree on which brands they recommend roughly two-thirds of the time (cross-provider recommendation Jaccard 0.35, below the 0.50-0.61 same-prompt rerun baseline). The picks diverge. But when neither provider recommends a brand, we classify the failure into one of three modes -- discoverability (the brand never reaches the model), compellingness (it reaches the model but isn't mentioned), or positioning (it's mentioned but not recommended) -- and on 7,763 such joint failures, both providers diagnose the same failure mode 95.1% of the time (clustered 95% CI [94.3%, 95.7%]). Agreement rises monotonically with falling brand prominence, from 81% [78.2%, 84.0%] on category leaders to 99.6% [99.3%, 9
Read on arXiv cs.AI →arXiv:2606.26117v1 Announce Type: cross Abstract: This paper introduces the Governance Inversion Hypothesis (GIH) to explain a growing paradox in artificial intelligence (AI) governance: under conditions of increasing regulatory expansion and technological complexity, organisations may become more formally governed while simultaneously experiencing a decline in operational control over AI systems. Existing AI governance frameworks generally assume that stronger regulation improves accountability, oversight, and organisational control. This paper challenges that assumption by arguing that governance formalisation itself may contribute to the erosion of control in AI-intensive environments. Drawing on institutional theory, organisational governance research, accountability scholarship, and emerging AI governance literature, the paper develops a conceptual framework explaining how regulatory expansion may weaken operational authority through four interconnected mechanisms: authority frag
Read on arXiv cs.AI →arXiv:2606.26183v1 Announce Type: cross Abstract: Building a generalist robot that can leverage prior knowledge for continuous task adaptation remains a significant challenge. Previous works alleviate the catastrophic forgetting problem by parameter-efficient fine-tuning for single-task adaptation. However, they fail to extract reusable skills and model the interaction with other skills effectively. Recent works try to address these issues by learning prompts. Differently, this paper presents an architectural perspective on the Lifelong Mixture of Dynamic Experts (\textit{LiMoDE}), a novel two-stage learning scheme for lifelong robot manipulation. Specifically, a dynamic MoE structure is first proposed in the multi-task pre-training stage to learn prior knowledge, where a varied number of heterogeneous experts are activated based on the motion information to address different short-term manipulations. Subsequently, in the task adaptation stage, we design a lifelong MoE adaptation mech
Read on arXiv cs.AI →FIFA is offering teams access to the same AI agent during this year's World Cup, raising questions about whether this approach ensures equal competition or if wealthier teams will still gain advantages through superior tools and resources.
Read on Wired AI →arXiv:2606.23702v1 Announce Type: cross Abstract: Modulation recognition systems rely on heterogeneous signal representations. 2D signal-image modalities such as time-frequency and cyclostationary maps capture structural patterns, while 1D statistical descriptors such as higher-order power spectra encode complementary cues. Under distribution shift, these modalities degrade unevenly, making robust fusion a central challenge for practical deployment. Progress is further limited by the lack of a unified evaluation protocol that systematically separates different shift types. This paper addresses both challenges through a joint benchmark-and-model study in underwater acoustic modulation recognition. UAMR-ShiftBench is the first benchmark to jointly cover in-distribution, low-SNR, unseen-environment, unseen-communication-parameter, and measured sea-trial evaluation under a single matched protocol, with two independent real-world subsets collected during two sea-trial campaigns conducted i
Read on arXiv cs.AI →arXiv:2606.23705v1 Announce Type: cross Abstract: Pain is assessed differently by patients, nurses, and clinicians, yet most computational approaches assume a single ground-truth label - effectively ignoring who is doing the rating. We introduce a rater-aware, event-aligned framework that converts sparse, rater-specific pain ratings into discrete pain-change events and aligns continuous wearable physiological signals to these events, preserving rater identity throughout. Applied to multimodal wearable data collected during spine-related pain procedures, the framework identifies substantial disagreement across rater groups and provides preliminary, exploratory evidence of rater-dependent physiological differences preceding reported pain increases. These findings suggest that pain-physiology relationships may not be rater-invariant, and that aggregating assessments across raters may mask meaningful physiological patterns. A rater-aware, event-aligned perspective is therefore a promising
Read on arXiv cs.AI →Meta released its own branded smart glasses without Ray-Ban partnership, available in multiple styles and colors including a collaboration with Kylie Jenner. The move marks Meta's shift toward independent smart glasses design after three years of partnering with Ray-Ban through EssilorLuxottica.
Read on The Verge AI →Amazon MGM has dropped director Luca Guadagnino's film about OpenAI CEO Sam Altman, which starred Andrew Garfield and dramatized Altman's brief termination and reinstatement in 2023. The studio stated it believes another distributor would be better suited to release the project.
Read on The Verge AI →Midjourney revealed its first hardware product, an ultrasound-based full-body scanner that uses sensor rings to capture internal body composition. The device aims to achieve image quality comparable to MRI and could be used for frequent health monitoring, representing a significant expansion beyond the company's AI image generation tools.
Read on The Verge AI →Google released a new $99.99 smart speaker featuring Gemini AI instead of traditional Google Assistant, enabling more natural conversational interactions rather than rigid command-based controls.
Read on TechCrunch AI →Google is releasing its first new smart speaker in six years, with preorders starting June 17th and shipping beginning June 29th. The $99 device features a rounded design with touch controls and a status light, available in four color options.
Read on The Verge AI →Google has released a new smart speaker featuring Gemini AI integration, marking its first major speaker hardware update in six years. The device is designed with Gemini's conversational capabilities as a core feature.
Read on Wired AI →WIRED and Architectural Digest have partnered to explore contemporary residential design and living patterns. The collaboration aims to provide insights into current home trends and future directions in how people inhabit spaces.
Read on Wired AI →Meta is rolling out an 'AI Mode' feature on Facebook that aggregates publicly available information from across Meta's platforms. The update represents the company's continued effort to integrate AI capabilities into its core social platform and boost user engagement.
Read on TechCrunch AI →Apple has introduced native AI photo editing tools in iOS 27 developer beta, marking the iPhone's first built-in AI editing capabilities. These features are less advanced than competitor offerings like Google Pixel but represent a significant shift in what iPhone's native photos app can do.
Read on The Verge AI →Andrew Yang identified residential costs, groceries, telecommunications, and other essentials as areas where Americans face inflated prices, positioning startups that reduce these expenses as the next major business opportunity.
Read on TechCrunch AI →A tech writer expresses ambivalence about personal AI assistants, questioning whether adopting such technology would create unhealthy dependency while acknowledging the appeal of AI-powered convenience features.
Read on TechCrunch AI →Alex Vindman, a former military officer who testified against Trump during the 2019 impeachment trial and subsequently left his military career, is now running for Senate. His candidacy represents a political challenge to Trump from someone with direct involvement in the earlier impeachment proceedings.
Read on Wired AI →A digital dentures project has concluded unsuccessfully. Details about the project's objectives, timeline, and reasons for failure are unavailable due to missing source material.
Read on Hugging Face →Hello Robot unveiled the fourth generation of its home assistance robot called Stretch. The California startup continues advancing domestic robotics technology for household use.
Read on TechCrunch AI →San Francisco Bay Area real estate listings are accepting Anthropic company stock as payment for homes, reflecting the AI startup's valuation and local wealth concentration. This arrangement highlights how private equity stakes have become alternative currency in the region's property market.
Read on Wired AI →arXiv:2606.02671v1 Announce Type: cross Abstract: Machine learning predictors have become essential tools for guiding automated decision making. However, a major misalignment persists: predictive models are typically optimized in terms of standard statistical metrics in isolation from the algorithmic tasks they inform. We highlight this incongruity in the high-stakes domain of organ allocation by demonstrating that any algorithm relying on (even highly accurate) survival predictors optimized for standard metrics -- such as the Concordance index (C-index) -- can yield arbitrarily poor outcomes when used for allocation, failing to guarantee utility better than a uniform random selection. To bridge the gap between survival analysis and policy optimization, we introduce a decision-focused learning approach based on optimizing normalized discounted cumulative gain (NDCG), a mainstay metric in information retrieval. We establish the utility of NDCG in survival analysis by proving that it tr
Read on arXiv cs.AI →arXiv:2606.02737v1 Announce Type: cross Abstract: Dense retrieval models exhibit positional bias: retrieval effectiveness degrades when relevant information appears later in a passage (Zeng et al., 2025). We ask whether this bias can be reduced at inference time, without retraining and without sacrificing overall retrieval effectiveness. To this end, we adapt inference-time attention calibration (Schuhmacher et al., 2026) to downstream retrieval and extend it with a strength coefficient lambda that interpolates between the original and fully calibrated attention distributions. Across three embedding models on SQuAD-PosQ and FineWeb-PosQ, we examine how basket size, calibrated layer set, and strength affect the trade-off between positional fairness and retrieval effectiveness, finding that partial calibration frequently outperforms full calibration. A single configuration (B=128, lambda=0.5, 50% layer depth) improves the harmonic mean of nDCG@10 across positional groups on FineWeb-PosQ
Read on arXiv cs.AI →Norse Atlantic Airways faces FTC complaints regarding its technology-focused customer service system, with multiple customers reporting significant financial losses. The budget airline's digital-first approach has raised consumer protection concerns.
Read on Wired AI →TechCrunch's Equity podcast episode examines whether technology executives show a particular tendency toward what some characterize as unrealistic or delusional thinking about AI's capabilities and near-term impact.
Read on TechCrunch AI →TechCrunch published a glossary explaining commonly-used AI terminology and jargon that has emerged as the field has grown. The resource helps readers understand key terms they may encounter in AI discussions and reporting.
Read on TechCrunch AI →Ferrari unveiled its first electric vehicle, the Luce, designed by Jony Ive with a distinctive futuristic aesthetic that diverges from Ferrari's traditional styling. The vehicle features innovative technology and design elements, though reactions from the public have been mixed with some strong criticism of its unconventional appearance.
Read on The Verge AI →Pope Leo cited The Lord of the Rings in an encyclical addressing AI, appearing to critique tech billionaires' misinterpretations of Tolkien's work. The religious leader's reference carries implicit commentary on how tech leaders have misused the literary work's themes.
Read on Wired AI →The Vatican invited Anthropic to participate in the presentation of Pope Leo's inaugural encyclical on AI, representing an unusual collaboration between the Catholic Church and a major AI company. The event signals the Church's engagement with Silicon Valley on ethical questions surrounding artificial intelligence technology.
Read on Wired AI →A person documented their daily household activities as a data collection method for training humanoid robots, exploring the trade-off between earning money and contributing personal behavioral data to AI development.
Read on Wired AI →An article discusses terminology related to AI agents, focusing on definitions of key concepts like 'harness' and 'scaffold' to ensure precise communication in the field. Proper understanding of these terms is important for developers and researchers working with AI systems.
Read on Hugging Face →A trial between Elon Musk and OpenAI leadership concluded with a jury dismissing all charges due to statute of limitations. Musk had sued Sam Altman and Greg Brockman, alleging OpenAI abandoned its nonprofit mission for profit, but the company countered that the lawsuit was a competitive attack motivated by Musk's rival AI ventures.
Read on The Verge AI →Google announced new subscription tiers for its AI services at I/O 2026, including a $100 monthly AI Ultra plan alongside updated features for existing Google AI Plus and Pro subscribers.
Read on Google AI Blog →OpenAI has added an optional safety feature to ChatGPT called Trusted Contact, which can alert a designated person if the system detects serious self-harm concerns during conversations.
Read on OpenAI →OpenAI outlined five guiding principles for its AGI development work, with CEO Sam Altman emphasizing the company's mission to ensure artificial general intelligence benefits everyone rather than concentrating power or resources.
Read on OpenAI →Hugging Face discusses a pull request scenario, though specific details about the change or its implications are not provided in the available information.
Read on Hugging Face →OpenAI announced a €500,000 grant program for European, Middle Eastern, and African organizations and researchers working on youth safety and wellbeing issues related to AI development and deployment.
Read on OpenAI →OpenAI is deploying an age prediction feature in ChatGPT to determine whether users are under or over 18 years old, enabling the application of age-appropriate safety measures. The system will be refined iteratively to improve its accuracy in age classification.
Read on OpenAI →NVIDIA announced DGX Spark and partnered with Reachy to deliver robotics and AI agent solutions. The initiatives aim to simplify deployment of autonomous systems and agents for enterprise applications.
Read on Hugging Face →Google DeepMind has announced a new project called Antigravity, though specific details about its purpose and capabilities were not provided in the available information.
Read on Google DeepMind →A development involving artificial intelligence applications for detecting or managing food allergies has been shared on the Hugging Face platform, though specific details about the implementation or breakthrough are not provided in the available information.
Read on Hugging Face →OpenAI is developing age prediction capabilities and parental controls for ChatGPT to tailor content appropriately for teenage users. The features aim to create safer experiences for younger users while providing families with additional oversight tools.
Read on OpenAI →OpenAI launches a $50M People-First AI Fund to help U.S. nonprofits scale impact with AI. Applications open Sept 8–Oct 8, 2025 for grants in education, healthcare, research, and more.
Read on OpenAI →OpenAI surveyed over 1,000 people worldwide on how AI should behave and compared their views to our Model Spec. Learn how collective alignment is shaping AI defaults to better reflect diverse human values and perspectives.
Read on OpenAI →OpenAI announced a new learning accelerator program, though specific details about its structure, eligibility, or benefits were not provided in the available information.
Read on OpenAI →Discover how MIXI, a leader in digital entertainment and lifestyle services in Japan, uses ChatGPT Enterprise to transform productivity, boost AI adoption across teams, and create a secure environment for innovation.
Read on OpenAI →Learn how Amgen uses GPT-5.
Read on OpenAI →We’re releasing gpt-oss-120b and gpt-oss-20b—two state-of-the-art open-weight language models that deliver strong real-world performance at low cost. Available under the flexible Apache 2.0 license, these models outperform similarly sized open models on reasoning tasks, demonstrate strong tool use capabilities, and are optimized for efficient deployment on consumer hardware.
Read on OpenAI →OpenAI is enhancing ChatGPT with features designed to support user wellbeing, including break reminders and improved guidance during difficult moments. The improvements are informed by expert consultation and aim to expand ChatGPT's utility beyond task completion to personal support.
Read on OpenAI →
Hi — I'm Scott. I'm building AI Info Forge because the Claude skills market is full of prompts that worked last month and break this month. Verified, drift-tested skills is the version of this market I wish existed. If you've felt the same friction, get on the list and I'll send you early access.
— Scott Sokolowski, Founder