All AI stories, newest first.
OpenAI has introduced a $100 monthly ChatGPT Pro subscription, offering five times the usage of its Codex coding tool compared to the Plus tier. This new tier is explicitly designed for professionals requiring extended, high-effort coding sessions.
OpenAI has released prompt-based safety policies for developers using gpt-oss-safeguard to moderate age-specific risks in AI systems. These guidelines aim to help developers create safer AI experiences for teenage users.
OpenAI and the Bill & Melinda Gates Foundation are collaborating to train disaster response teams in Asia to leverage AI for more effective emergency management. The initiative aims to bridge the gap between AI capabilities and real-world disaster response needs.
OpenAI has acquired TBPN to foster global discussions about AI and support independent media. The move aims to engage builders, businesses, and the tech community more deeply.
OMLX is a new tool designed to optimize large language model inference specifically for Mac users. It promises to make running LLMs on Apple Silicon faster and more efficient.
NVIDIA has released a tool to fine-tune domain-specific embedding models quickly. This democratizes access to specialized AI applications.
Nous Research's NousCoder-14B matches proprietary systems in coding tasks, trained in just four days. This release comes as Claude Code dominates developer discussions.
A new study explores how emotions affect decision-making in small language models (SLMs) by using activation steering for controlled emotional states. The research introduces a game-theoretic benchmark to evaluate these interactions.
Researchers propose a novel framework treating LLM hallucinations as output-boundary misclassification errors, introducing a composite abstention architecture. This system combines instruction-based refusal with a structural gate that blocks unsupported claims based on a calculated support deficit score.
Researchers developed an LLM-based tool to identify HIV-related stigma in clinical narratives, addressing a critical gap in healthcare documentation. The tool could improve mental health outcomes and treatment adherence for people living with HIV.
Researchers introduce Contextual Earnings-22, a benchmark highlighting the gap between academic and real-world speech recognition performance. The study emphasizes the importance of contextual conditioning in high-stakes domains.
A new type of computer is emerging, blending neural networks with traditional hardware. This hybrid approach promises to revolutionize computing by merging AI and classical computing paradigms.
Netflix has implemented a large language model to evaluate and generate show synopses, improving content discovery. This marks a significant step in AI-driven content curation for streaming platforms.
Meta is leveraging AI to defend against social-media lawsuits, aiming to reduce legal risks. The move highlights the growing role of AI in legal strategies and corporate defense.
Meta's AI app has skyrocketed from No. 57 to No. 5 on the App Store following the release of its new Muse Spark model. This rapid ascent highlights the growing demand for advanced AI tools among consumers.
Google's MedGemma 1.5 4B model now integrates high-dimensional medical imaging, anatomical localization, and multi-timepoint analysis within a single architecture. This update significantly expands capabilities to include CT/MRI volumes, histopathology, and complex EHR document understanding.
Researchers propose a framework using LLMs to validate and restructure unsupervised text clusters, improving coherence and reducing redundancy. The method leverages LLMs as semantic judges rather than embedding generators.
Listen Labs secured $69 million in funding after a creative billboard campaign that attracted top engineering talent. The campaign featured AI tokens leading to a unique coding challenge, bypassing traditional hiring methods.
The Linux kernel now permits the use of AI tools for contributions, with strict guidelines to ensure code quality. This marks a significant shift in open-source development practices.
Researchers propose a new method for resource-aware knowledge distillation in multi-agent reinforcement learning (MARL). The approach addresses the challenges of deploying expert policies on edge devices by focusing on coordination structure and heterogeneous agent capabilities.