AI News Today

Article URL: https://ffilm.org/astro/ Comments URL: https://news.ycombinator.com/item?id=48930761 Points: 3 # Comments: 0

<pre style='white-space:pre-wrap;width:81ex'>fix(sessions): stop leaking file path as prompt content on read failure (#108546) * fix(sessions): stop leaking file path as prompt content on read failure When readFileSync fails for a valid file path, resolvePromptInput returns the raw path string as prompt content instead of undefined. This injects filesystem paths into the LLM context. The existing console.error warning still fires; the caller already handles undefined returns correctly. Co-Author

<pre style='white-space:pre-wrap;width:81ex'>fix(agents): cleanup leaked toolStartData entries on unsubscribe (#108403) * fix(agents): cleanup leaked toolStartData entries on unsubscribe toolStartData module-level Map only deleted entries on handleToolExecutionEnd; crashed/aborted/unsubscribed tool executions left leaked entries, inflating countActiveToolExecutions. Ref: BUG-043 (local finding) Co-Authored-By: Claude <claude@anthropic.com> * test(agents): cover tool start cleanup on unsubscribe

Article URL: https://www.ode.com Comments URL: https://news.ycombinator.com/item?id=48930726 Points: 2 # Comments: 2

<p>AI answer engines — Google's AI Overviews, ChatGPT Search, Perplexity, Gemini — increasingly answer users without a click. For developers, that shifts the goal: it's no longer only "rank the page," it's "be the source the model quotes." The good news is that a lot of what makes a site citable is squarely front-end and infra work. Here's the checklist my team actually uses.</p> <h2> 1. Server-render anything you want cited </h2> <p>LLM crawlers are far better at parsing HTML that exists <stron

<p>Razorpay trips people up in a way most providers do not: it has two separate signatures that use the same algorithm but different keys and different inputs, and they are easy to confuse. One verifies webhooks. The other verifies a checkout payment on your success handler. Sign the wrong thing with the wrong secret and your verification fails while your code looks correct. This post shows the exact Razorpay webhook verification, the raw-body trap, and how the webhook signature differs from the

<p><em>This is a submission for <a href="https://dev.to/bugsmash">DEV's Summer Bug Smash: Clear the Lineup</a> powered by <a href="https://sentry.io/" rel="noopener noreferrer">Sentry</a>.</em></p> <h2> Project Overview </h2> <p>I picked <a href="https://github.com/huggingface/smolagents" rel="noopener noreferrer">huggingface/smolagents</a>, the 28k+ star agent framework where agents literally think in Python code. It had a bug open since <a href="https://github.com/huggingface/smolagents/issues

Article URL: https://www.youtube.com/watch?v=6zwuTqGweJE Comments URL: https://news.ycombinator.com/item?id=48930685 Points: 5 # Comments: 1

Article URL: https://steelsoft.site/software/DeepCleanAudio_home.html Comments URL: https://news.ycombinator.com/item?id=48930676 Points: 2 # Comments: 0

<p>B2B software applications must provide security audits to sell to enterprise customers. When moving upmarket, corporate buyers require a detailed history of sensitive events. They want to know exactly who did what, when it happened, and what data changed. Meeting these demands is essential to satisfy compliance rules like SOC 2 and ISO 27001, as well as the EU AI Act.</p> <p>However, software developers face a split in the tooling market. Some tools are built for enterprise compliance, while

<p> Turn your camera roll into social content with AI </p> <p> <a href="https://www.producthunt.com/products/clipmatch?utm_campaign=producthunt-atom-posts-feed&utm_medium=rss-feed&utm_source=producthunt-atom-posts-feed">Discussion</a> | <a href="https://www.producthunt.com/r/p/1197826?app_id=339">Link</a> </p>

<p> Your teachable AI intern for everyday browser work </p> <p> <a href="https://www.producthunt.com/products/aye-browser?utm_campaign=producthunt-atom-posts-feed&utm_medium=rss-feed&utm_source=producthunt-atom-posts-feed">Discussion</a> | <a href="https://www.producthunt.com/r/p/1197816?app_id=339">Link</a> </p>

Comments

Article URL: https://tpmme.me Comments URL: https://news.ycombinator.com/item?id=48930345 Points: 1 # Comments: 0

Hey Guys, I've built one of the largest Open Source AI starter templates using Next JS and OpenAI tech stack. It has a wide range of fully functioning applications from Voice AI agents to multimodal Chatbots. Any feedback would be great: https://suhasbhairav.com/ai-templates Regards Suhas Bhairav Comments URL: https://news.ycombinator.com/item?id=48930314 Points: 1 # Comments: 1

Article URL: https://www.wsj.com/us-news/the-ai-backlash-has-tech-executives-fearing-for-their-lives-30c43972 Comments URL: https://news.ycombinator.com/item?id=48930281 Points: 16 # Comments: 8

Article URL: https://www.economist.com/leaders/2026/07/15/how-to-make-ai-safe-and-free-of-america-and-china Comments URL: https://news.ycombinator.com/item?id=48930223 Points: 4 # Comments: 0

Article URL: https://arnoldkling.substack.com/p/there-is-no-ai-in-team Comments URL: https://news.ycombinator.com/item?id=48930216 Points: 4 # Comments: 0

Built an desktop AI companion that always pinned at top. Some of the features: - Lightweight: Built on Tauri v2 + React + Rust for low RAM usage. - Always-on-Top Floating Mode: Pinned window stays on top so you can code and query AI side-by-side (no more app-switching!). - Secure Keyring Storage: API keys are saved directly in your OS secure Keychain/Keyring - Local SQLite Database: All chat history and threads are saved locally on your device. - Multi Provider Support: Connect to Gemini, OpenAI

<p>Uncrewed systems are the future for armed forces and the Netherlands is leading the way ‘to keep people out of danger zones’</p><p>On each side of the target ship, a black vessel keeps a watchful distance. Defender 1 and Defender 2 are the eyes and ears of the navy – but they have nobody onboard, and their paths are controlled by a computer system.</p><p>This is the future of the Royal Netherlands Navy, according to Capt Sjoerd Feenstra, head of the expertise centre for unmanned systems. He i

Researchers developed an automated framework that helps AI models generate CAD programs more accurately and efficiently.

Applied Computing has raised a $20M Series A to build a foundation AI model for the oil, gas and petrochemical industry.

arXiv:2607.13037v1 Announce Type: new Abstract: When a data contributor requests removal, model trainers face a practical gap: unlearning algorithms require a forget set, yet no tool can locate which training records belong to a given author. Existing provenance systems operate at file or dataset level, forcing catastrophic over-deletion. We present ob, a record- and token-level data provenance system that propagates author identity through data processing pipelines and resolves revocation reque

arXiv:2607.13049v1 Announce Type: new Abstract: Foundation models have given robots a sophisticated brain for complex decision-making, yet deploying that intelligence into a physical platform still demands tedious, expert-driven calibration. This deployment gap, the robot's spinal cord, remains a primary bottleneck to scalable Embodied AI. Hence, we propose SPINE (Scalable Physical Integration with ageNtic Expertise): an agentic framework for systematically debugging and deploying bimanual robot

arXiv:2607.13069v1 Announce Type: new Abstract: Large language models produce chain-of-thought (CoT) reasoning that appears logically sound yet may not genuinely depend on its stated premises. We introduce interventional grounding audits, a black-box, step-level test of premise dependency: we intervene on a single premise by substituting its target predicate with a fresh symbol, re-run the model, and check whether each reasoning step's normalized conclusion (canonical predicate form) changes. We

arXiv:2607.13073v1 Announce Type: new Abstract: Neuro-symbolic AI based on $IFOL_B$ is a way to combine neural learning and symbolic reasoning to overcome limitations of purely neural systems (like lack of interpretability and logical structure) with formal logical machinery for self-reference. In this paper we expand the cognitive power of $IFOL_B$ by using the probability computation for the currently unknown sentences, based on Nilsson's probability structure for the $IFOL_B$. We introduce th

arXiv:2607.13104v1 Announce Type: new Abstract: Self-improving autonomous agents are moving from research prototypes to deployed systems. The primary goal is controllable evolution, or adaptation, from experience with minimal or even no human input. This survey frames modern self-improving agents as adaptive systems that convert experience into accumulated capability gains. We offer a system-level framework that represents a modern agent as a configuration coupling a foundation model with an ope

arXiv:2607.13115v1 Announce Type: new Abstract: Small language models (SLMs) have shown promise for zero-shot molecular property prediction from SMILES strings, yet they often suffer from structural blindness because sequence representations under-specify key graph-topological cues. We propose a modular Context-Augmented Prompting framework that enables agentic tool use at inference time: a trained GNN expert model provides a predictive hint with confidence, and a GNN extracts an instance-specif

arXiv:2607.13157v1 Announce Type: new Abstract: Agent memory is a systems problem for long-horizon agents. Practical deployments require retention of task state across extended conversations, recovery of user-specific facts and preferences across sessions, and accumulation of procedural knowledge from prior outcomes. These requirements extend beyond document retrieval: a memory layer must determine which interactions become durable state, how that state is scoped, how it is retrieved under laten

arXiv:2607.13172v1 Announce Type: new Abstract: We address the problem of safely training an agent policy and deploying a good and safe policy, in settings where the environment dynamics are unknown and no suitable reward function is available. In the context of safety-critical environments, we consider traditional reinforcement learning impractical and resort to the resource of human input. We introduce DROPJ, a human-centred method for both safe training and deployment. We first learn a world

arXiv:2607.13219v1 Announce Type: new Abstract: We present cayleyR, an R package for solving permutation puzzles by detecting cycle intersections in Cayley graphs. The core algorithm performs an iterative bidirectional search: from both the initial and target permutation states, random operation sequences generate cycles in the Cayley graph of the symmetric group Sn; their intersection yields a connecting path. When no direct intersection is found, a distance-guided bridge selection narrows the

arXiv:2607.13220v1 Announce Type: new Abstract: Most AI-for-science systems focus on scaling a single reasoning process through better models, larger context windows, long-horizon agentic execution, or digital co-scientists working with one principal user. However, challenging scientific problems are rarely solved by one reasoner alone. They are solved by teams whose members bring different priors, experimental backgrounds, tacit knowledge, and domain-trained intuitions. The open problem is ther

arXiv:2607.13230v1 Announce Type: new Abstract: Agentic AI introduces new insurance challenges because autonomous AI systems can make decisions, invoke tools, modify external environments, and interact with third-party services. This paper develops an AI-native mathematical framework for underwriting, pricing, and contract design for agentic AI deployments. A deployment is represented by a risk state that captures autonomy level, operational authority, permission exposure, governance maturity, a

arXiv:2607.13239v1 Announce Type: new Abstract: Foundation models, including large language models (LLMs) and vision-language models (VLMs), are increasingly used for transportation management center (TMC) tasks such as anomaly detection, incident reporting, and traveler information. Deploying multiple such models across TMC functions raises a portfolio question: which model should serve each function, in which deployment mode, and under what shared hardware budget? We formulate this as the Foun

arXiv:2607.13285v1 Announce Type: new Abstract: The capability of a modern AI agent depends not only on its foundation model but also on its harness, which constructs prompts, manages state, invokes tools, and coordinates execution. As models, APIs, environments, and requirements evolve, the harness must be continually modified. Before such a change can be made, a developer or coding agent must identify all code locations that implement the target behavior. This is difficult because production h

arXiv:2607.13396v1 Announce Type: new Abstract: What happens to an LLM agent's tool choice when the reliable tool silently changes within an ongoing session? We borrow set-shifting from cognitive psychology to study how well agents adapt to hidden reliability shifts. Our benchmark mounts tool-skill libraries with redundancies, where many tools solve the same task but differ in hidden reliability. In our evaluation framework, a branched schedule shifts the reliable tool group at hidden boundaries

arXiv:2607.13548v1 Announce Type: new Abstract: Identifying root causes in production microservice failures requires reasoning over large-scale, multimodal telemetry spanning metrics, logs, and traces, a problem that has proved resistant to both classical and LLM-based approaches. The OpenRCA dataset exemplifies these challenges: it is large-scale, multimodal, and lacks detailed domain knowledge, and yields consistently low accuracy across all existing methods. We show that classical causal disc

arXiv:2607.13558v1 Announce Type: new Abstract: Urban region profiling constitutes a core problem in urban computing, supporting applications such as population estimation, economic assessment, and environmental monitoring. Existing methods typically formulate this task as multimodal representation learning, fusing heterogeneous urban data, e.g., satellite imagery, points of interest, textual descriptions, and 3D building information, into latent embeddings for prediction. However, these approac

arXiv:2607.13562v1 Announce Type: new Abstract: Knowing when to say "I don't know" is fundamental to human judgment, yet AI assistants offer a fluent answer to almost any question. In five experiments (N = 3,132; four preregistered, one direct replication), participants answered difficult questions and could always decline to respond. We engineered the questions so that AI advice was wrong, separating AI use from its accuracy. Merely having access to AI nearly eliminated participants' willingnes

arXiv:2607.13594v1 Announce Type: new Abstract: LLM agents act on real-world environments through tool calls, and a single misjudged action can cause irreversible harm. The standard safeguard is a guard model that labels each proposed action as safe or unsafe, but this binary view conflates two distinct decisions: whether the action is harmful in itself, and whether it is appropriate given the user's context. It also operates at the granularity of action categories rather than individual instanc