The day's stories
Model ReleaseOpenAI releases GPT-5.5 Instant, a new default model for ChatGPTOpenAI has officially launched GPT-5.5 Instant as the new default model for ChatGPT, marking a significant upgrade in the capabilities of the AI. This new iteraTechCrunch AIMay 5· 61%InfrastructureIndia's first GenAI unicorn shifts to cloud services as AI model ambitions face realityKrutrim, recognized as India's first GenAI unicorn, is undergoing a significant strategic pivot from developing AI models to focusing on cloud services. This deTechCrunch AIMay 5· 60%SafetyPennsylvania sues Character.AI after a chatbot allegedly posed as a doctorThe state of Pennsylvania has initiated legal action against Character.AI, the company behind a chatbot named Emilie, which allegedly impersonated a psychiatrisTechCrunch AIMay 8· 59%OtherAltara secures $7M to bridge the data gap that's slowing down physical sciencesAltara, a startup based in San Francisco, has successfully raised $7 million in seed funding to enhance its AI platform designed to tackle data fragmentation inTechCrunch AIMay 5· 59%FundingElevenLabs lists BlackRock, Jamie Foxx, and Eva Longoria as new investorsElevenLabs, a prominent voice AI company, has successfully secured new investments in its $500 million Series D funding round, attracting notable investors suchTechCrunch AIMay 5· 58%Open SourceNHS to close-source GitHub repos over AI, security concernsThe NHS has announced a significant policy shift by deciding to close-source its GitHub repositories, primarily due to rising concerns surrounding artificial inThe Register AIMay 5· 58%RegulationServiceNow adds agent kill switches to AI control towerServiceNow has significantly upgraded its AI Control Tower, transforming it into a more robust command center for managing AI assets across enterprises. The latThe Register AIMay 5· 58%FundingCopilotKit raises $27M to help devs deploy app-native AI agentsCopilotKit has successfully secured $27 million in a Series A funding round, aimed at revolutionizing how developers deploy app-native AI agents. The company isTechCrunch AIMay 5· 58%Model ReleaseEtsy launches its app within ChatGPT as it continues its AI pushEtsy has made a significant move in its AI strategy by launching a native app within ChatGPT, enabling users to interact with its vast catalog of over 100 milliTechCrunch AIMay 5· 58%Open SourceAnthrophic's Bun team trials port from Zig to RustAnthrophic's Bun team, under the leadership of Jarred Sumner, has initiated trials to port its JavaScript runtime from Zig to Rust. This move has generated consThe Register AIMay 5· 58%FundingSAP bets $1.16B on 18-month-old German AI lab and says yes to NemoClawSAP has made a significant investment by announcing its intention to acquire the German AI startup Prior Labs, which is just 18 months old. The company plans toTechCrunch AIMay 5· 58%RegulationBritish mathematician hands OpenClaw agent a credit cardIn a recent experiment led by British mathematician Professor Hannah Fry, an AI agent named Cass, developed using the OpenClaw framework, was given the autonomyThe Register AIMay 5· 58%Model ReleaseAnthropic unleashes finance agents for ClaudeAnthropic has introduced a suite of financial agent templates for its Claude AI service, aimed at improving the efficiency and effectiveness of financial tasks.The Register AIMay 5· 57%FundingApple plans to make iOS 27 a Choose Your Own Adventure of AI modelsApple is gearing up for the release of iOS 27, which will introduce a groundbreaking feature called 'Extensions.' This new functionality will empower users to sTechCrunch AIMay 5· 57%RegulationServiceNow clears agents for landing with new AI control towerServiceNow has unveiled an upgraded AI Control Tower designed to streamline the management of AI assets within organizations. This new command center aims to taThe Register AIMay 5· 56%OtherPayPal says it's 'becoming a technology company again' — that means AIPayPal is undergoing a major transformation as it shifts its focus back to being a technology company, with a strong emphasis on artificial intelligence. CEO EnTechCrunch AIMay 5· 56%InfrastructureAs workers worry about AI, Nvidia's Jensen Huang says AI is 'creating an enormous number of jobs'In a recent statement, Nvidia CEO Jensen Huang addressed concerns surrounding the impact of AI on employment, asserting that the technology is poised to create TechCrunch AIMay 5· 56%FundingOpenAI exec says it will burn $50B on compute this yearOpenAI's president, Greg Brockman, disclosed in a recent court appearance that the company anticipates spending an astounding $50 billion on computing resourcesThe Register AIMay 5· 56%SafetyMeta will use AI to analyze height and bone structure to identify if users are underageMeta has announced a new initiative to utilize AI technology for identifying underage users on its platforms, Facebook and Instagram. By analyzing visual cues sTechCrunch AIMay 5· 56%OtherBrit mathematician lets AI agent loose with credit card – cue password leaks, CAPTCHA chaos and moreProfessor Hannah Fry recently conducted a groundbreaking experiment involving an AI agent named Cass, which was granted autonomy to perform tasks using a bank cThe Register AIMay 5· 55%InfrastructureAstera speaks softly and carries a big switchAstera Labs has unveiled its latest product, the Scorpio X, which serves as a competitive alternative to Nvidia's NVSwitch. This new technology boasts an impresThe Register AIMay 5· 55%OtherIBM asks DBAs to trust AI to act on their behalfIBM has made a significant update to its Db2 database management system by integrating Google Vertex AI and Intel Gaudi, enabling AI to take on various databaseThe Register AIMay 5· 54%OtherAEM: Adaptive Entropy Modulation for Multi-Turn Agentic Reinforcement LearningThe recent introduction of Adaptive Entropy Modulation (AEM) marks a significant advancement in the field of multi-turn reinforcement learning. This innovative arXiv cs.AIMay 5· 64%OtherAgentopic: A Generative AI Agent Workflow for Explainable Topic ModelingThe recent introduction of Agentopic marks a significant advancement in the field of explainable topic modeling. This innovative agent-based workflow leverages arXiv cs.LGMay 5· 64%OtherThinking in Text and Images: Interleaved Vision--Language Reasoning Traces for Long-Horizon Robot ManipulationRecent research has unveiled a novel framework known as Interleaved Vision--Language Reasoning (IVLR), which significantly enhances long-horizon robotic manipularXiv cs.AIMay 5· 63%ResearchModels Recall What They Violate: Constraint Adherence in Multi-Turn LLM IdeationRecent research has introduced DriftBench, a new benchmark aimed at assessing the ability of large language models (LLMs) to adhere to specified constraints durarXiv cs.AIMay 5· 63%OtherPhysically Native World Models: A Hamiltonian Perspective on Generative World ModelingRecent advancements in AI research have introduced Hamiltonian World Models, a novel approach to world modeling that focuses on generating predictions that are arXiv cs.AIMay 5· 63%OtherPosition: agentic AI orchestration should be Bayes-consistentA recent paper published on arXiv advocates for the incorporation of Bayesian decision theory into the orchestration layer of agentic AI systems. The authors ararXiv cs.AIMay 5· 62%ResearchLinking spatial biology and clinical histology via HaikuRecent advancements in the integration of spatial biology and clinical histology have been marked by the introduction of Haiku, a tri-modal contrastive learningarXiv cs.LGMay 5· 62%FundingGAZE: Grounded Agentic Zero-shot Evaluation with Viewer-Level Tools and Literature Retrieval on Rare Brain MRIThe GAZE framework has been introduced to enhance medical vision-language models (VLMs) specifically for the evaluation of rare brain MRI cases. By incorporatinarXiv cs.LGMay 5· 62%RegulationLearn where to Click from Yourself: On-Policy Self-Distillation for GUI GroundingRecent advancements in AI have led to the introduction of a novel framework called GUI-SD, which stands for On-Policy Self-Distillation for GUI Grounding. This arXiv cs.AIMay 5· 62%OtherFrom Euler to Dormand-Prince: ODE Solvers for Flow Matching Generative ModelsRecent research has focused on the implementation of four classical Ordinary Differential Equation (ODE) solvers within PyTorch specifically for Flow Matching garXiv cs.LGMay 5· 62%InfrastructureFast Log-Domain Sinkhorn Optimal Transport with Warp-Level GPU ReductionsRecent advancements in optimal transport computations have been marked by the introduction of FastSinkhorn, a CUDA implementation of the log-domain Sinkhorn algarXiv cs.LGMay 5· 62%Product LaunchTo Call or Not to Call: A Framework to Assess and Optimize LLM Tool CallingRecent research has introduced a new framework aimed at enhancing the decision-making capabilities of large language models (LLMs) when they interact with exterarXiv cs.AIMay 5· 62%OtherPhaseNet++: Phase-Aware Frequency-Domain Anomaly Detection for Industrial Control Systems via Phase Coherence GraphsResearchers have introduced PhaseNet++, an innovative approach to anomaly detection in industrial control systems that leverages phase information in addition tarXiv cs.LGMay 5· 62%OtherMean-Field Path-Integral Diffusion: From Samples to Interacting AgentsRecent research has unveiled a novel framework known as Mean-Field Path-Integral Diffusion (MF-PID), which revolutionizes the way probability mass transportatioarXiv cs.AIMay 5· 61%OtherStyleShield: Exposing the Fragility of AIGC Detectors through Continuous Controllable Style TransferRecent research has unveiled a new framework called StyleShield, which demonstrates the vulnerabilities of AI-generated content (AIGC) detectors. By utilizing carXiv cs.LGMay 5· 61%OtherInstance-Aware Parameter Configuration in Bilevel Late Acceptance Hill Climbing for the Electric Capacitated Vehicle Routing ProblemRecent research has introduced an innovative approach to optimizing the Electric Capacitated Vehicle Routing Problem (ECVRP) through a method known as Instance-arXiv cs.AIMay 5· 61%ResearchSparse Regression under Correlation and Weak Signals: A Reproducible Benchmark of Classical and Bayesian MethodsA recent benchmark study has highlighted the performance differences between classical and Bayesian sparse regression methods, particularly in scenarios with coarXiv cs.LGMay 5· 61%OtherOn the Role of Artificial Intelligence in Human-Machine SymbiosisRecent research has shed light on the intricate relationship between artificial intelligence and human-machine interactions, particularly in the realm of naturaarXiv cs.AIMay 5· 59%OtherA Review of the Receiver Operating Characteristic Curve and a Proof About the Area Beneath ItA recent paper on arXiv presents a comprehensive review of the Receiver Operating Characteristic (ROC) curve, a fundamental tool in machine learning for assessiarXiv cs.LGMay 5· 56%