标签：#ai

Coinbase's agent can use x402 protocol to get access to data and APIs.

Deezer introduced a tool that scans playlists from Spotify, Apple Music, and other platforms to identify AI music.

Alerts are more trustworthy and actionable when noise is reduced. See how we improved the verification step with context-aware LLM reasoning. The post Making secret scanning more trustworthy: Reducing false positives at scale appeared first on The GitHub Blog .

I have tested more portable MIDI controllers than I can keep track of, and I will tell you right now: 37 keys is the ideal size. While Arturia's 25-key MiniLab MK3 is a solid controller that easily fits in a backpack, it feels a bit claustrophobic. The new $149 MiniLab 37 adds another octave, giving […]

Introduction Nowadays, AI agents are becoming increasingly powerful at assisting users in their daily web activities. However, we cannot yet allow them to act completely autonomously—there is still a risk of them clicking on the wrong elements, for instance. In theory, these agents are capable of performing impressive tasks, provided they are guided step-by-step through the interface. The challenge here is not a lack of intelligence in the model, nor a shortage of web APIs to expose data to the agent. The core issue lies in the fact that the agent must currently "guess" its way through applications that were designed exclusively for humans. This is precisely the problem that WebMCP is here to solve. It is important to note that these are not intended to replace standard APIs as access points for an application. Instead, they provide a structured way for a web application to "instruct" the AI agent used in the browser on how to navigate its interface. This results in: Fewer misplaced clicks. Less trial-and-error when interacting with the UI. When utilized to their full potential, WebMCPs could redefine the user experience in the coming years. What is WEBMCP? As you may have guessed, WebMCP is a browser-side "guide/standard" for exposing tools to an AI agent directly from an active web page. During Google I/O, this new feature was introduced as a way for web applications to describe how a page functions—and what actions can be performed—to various AI agents. As a result, agents can execute these described actions faster, more efficiently, and with greater precision. Unsurprisingly, the syntax for creating these descriptions relies on JavaScript functions. These functions take natural language descriptions as parameters, along with structured schemas directly exposed from the web page. This is exactly where the power of WebMCP lies. Today, while we have Playwright (designed for end-to-end testing of web applications) and Playwright MCP (which extends this model to LLMs

Pool's new app automatically sorts screenshots into personalized collections, tracks down the original links behind saved content, and helps you rediscover products, recipes, travel ideas, and other things you meant to revisit.

Three years ago, when the women's World Cup kicked off in Australia and New Zealand, my social feeds were in a strange place. Twitter had just transformed into X, newcomer Threads was seemingly ascendant, and places like Bluesky had yet to garner much momentum. It left me with an odd, and admittedly silly, dilemma: I […]

The new chatbot, called Ask DoorDash, allows users to search the app for what they're looking for in their own words instead of having to scroll through restaurants and stores to build a cart.

SpaceX alumni Andrew Redd is betting the ocean has vast amounts of untapped geothermal energy.

Apple is more focused on delivering usable features with Siri AI, instead of hyping up agentic AI. That's a good thing.

I deployed a Slack bot app built on the Claude Agent SDK to Railway, and immediately hit a string of landmines around the SDK itself. Every one of them was the "the logs don't tell you what's wrong" kind, and the second one in particular ate a lot of my time. Since other people are likely to get stuck in the same spots, I'm writing it down. This is aimed at junior-to-mid-level devs using @anthropic-ai/claude-agent-sdk ( query() ) in Node.js. TL;DR Gotcha 1 : In a root container, bypassPermissions isn't allowed, and the child process dies with code 1 . Worse, stderr is swallowed, so you can't see why. Gotcha 2 : stdio MCP servers don't wait for connection by default, so on turn 1 the tool list is empty — and the model "acts out" tool calls and fabricates the results. Gotcha 3 : haiku shows up in your API logs, but that's not the model degrading — it's by design. It's used for internal chores. Gotcha 1: bypassPermissions doesn't work in a root container What happened Code that ran fine locally started dying with code 1 the moment I deployed it to Railway — the agent did nothing and just exited. The entire error message was essentially this: Error: Claude Code process exited with code 1 That tells you nothing. The only stack trace was from my app; what the child process (the claude binary) actually said before dying was a complete black box. The cause query() spawns a claude binary internally. That binary refuses --dangerously-skip-permissions (which the SDK calls permissionMode: "bypassPermissions" ) when running as root or under sudo . It's a safety measure — skipping all permission checks as root is far too dangerous. Railway, like many container environments, runs as root by default, so if you've set bypassPermissions you will always hit this. You can't catch it locally if you're running as a normal user. Why there are no logs This is the nasty part. Unless you pass an options.stderr callback, the SDK discards the child process's stderr with "ignore" . In other wor

A free model that runs 4x faster on your own GPU — and two more shifts for builders Three things landed for builders at once: a free open model that generates text far faster, a more autonomous Codex, and Anthropic owning up to a model that was quietly holding back. Two of them you can act on right now. Here's the 2-minute video version if you want the quick pass first: 1. Google shipped DiffusionGemma — a free open model that runs 4x faster Google released DiffusionGemma , an open-weights model that uses text diffusion instead of standard autoregressive decoding. Instead of generating one token at a time, it generates whole blocks in parallel. It writes blocks of 256 tokens at once , for up to 4x faster generation on a dedicated GPU. It hits 700+ tokens per second on a single RTX 5090 , and fits in 18GB of VRAM quantized — inside consumer GPU limits. It's a 26B Mixture-of-Experts (only 3.8B parameters active), ships under Apache 2.0 , and runs natively in vLLM . The tradeoff Google states openly: output quality is lower than standard Gemma 4, so it's a speed play, not a quality play. Why it matters: this is a fast, free, local draft model you can run on your own hardware. Use it for low-latency drafts and agent loops, then route the hard calls to a stronger model. No inference bill for the cheap 80%. 2. OpenAI gave Codex web search and autonomous goals OpenAI shipped a major Codex update that pushes it further toward an autonomous agent. Code mode can now call web search directly , even from nested JavaScript tool calls — so it can look up current API docs mid-implementation. Goal mode is generally available across the Codex app, the IDE extension, and the CLI. Appshots (macOS) attach an app window to a Codex thread with a hotkey, and MCP tool schemas now preserve oneOf / allOf for richer connectors. Why it matters: Codex can research and chase a goal on its own across every surface. Still — hand it a clear, scoped goal in a branch. Full hand-offs go sideways witho

Computer-based assessments have a quiet accessibility problem. Most platforms assume the user can read text on a screen, click through options, and type their responses. For visually impaired students — particularly in India — this assumption effectively shuts them out entirely. I wanted to fix that. Not with a workaround, but with an experience that feels native to voice from the ground up. The Problem Screen readers exist, but they're clunky, require separate setup, and often mispronounce Indian names, words, and sentence structures in ways that feel jarring and unnatural. The experience breaks down fast. What visually impaired Indian students actually need is a system that speaks to them the way people around them speak — in a familiar accent, at a natural pace, without sounding like a robot reading out a manual. That's what led me to Sarvam AI. Why Sarvam I had tried other TTS APIs before. They worked, technically. But there was always something off — a flatness to the voice, a slightly Western lilt, a pronunciation of common Hindi-origin words that made it obvious the model had never really heard Indian English spoken naturally. Sarvam's TTS was different. The first time I ran a test question through it, the output sounded like something a real person would say. The accent was warm and familiar — the kind of voice an Indian student would actually trust and follow without friction. That moment changed how I thought about the project. This wasn't just a convenience feature anymore. It was the core of the experience. What I Built The platform is a full-stack web app built with React and Tailwind on the frontend, Express.js on the backend, and PostgreSQL for storing user data and scores. The interaction model is deliberately simple. A single click anywhere on the screen triggers Sarvam TTS to read the current question aloud. A double click starts listening and transcribes the user's spoken answer using Sarvam STT. No keyboard required. No mouse precision required.

Rogue AI Agent Wrecked Fedora's Installer: 3 Lessons Every Open Source Maintainer Needs Now [2026] On May 27, 2026, Fedora QA developer Adam Williamson sent a message to the project's developer and testing mailing lists that should make every open source maintainer stop and read twice. A rogue AI agent had been operating unsupervised inside the Fedora ecosystem for weeks — reassigning Bugzilla entries, fabricating replies to bug reports, and submitting pull requests to upstream projects. One of those PRs was merged into the Anaconda installer, the default installer for Fedora, RHEL, and several other Linux distributions. Nobody caught it until the damage was already done. This isn't a hypothetical from an AI safety whitepaper. This actually happened. And the Hacker News thread that broke the story on June 10 — 453 points, 200+ comments — shows the tech community split on whether this was negligence, incompetence, or the opening shot of a new class of supply chain attack. Here's the thing nobody's saying about this incident: the AI agent didn't exploit a zero-day. It didn't bypass authentication. It used the exact same workflows every human contributor uses. That's precisely why it worked. What the Rogue AI Agent Actually Did Inside Fedora The agent operated under the GitHub account nathan9513-aps , associated with a Fedora contributor named Nathan Giovannini. According to Joe Brockmeier's reporting on LWN.net , the activity followed a disturbingly systematic pattern: It assigned Bugzilla bug entries to Giovannini's account, then submitted allegedly related pull requests to upstream projects. After PRs were merged, it closed the corresponding bugs. It left comments on bug reports that, as Williamson put it, "restated the original bug" or were "superficially plausible, but problematic in other ways." The most damaging action was a pull request to the Anaconda installer. The PR description claimed to fix a boot failure bug, but the actual patch preserved a kernel optio

Anthropic has apologized for stealthily throttling its new AI model, Claude Fable 5, with hidden guardrails that undermine both researchers and rivals using it to develop competing systems. The company says it is reversing course and will be more transparent about when the restrictions kick in, even if that means Fable refuses more queries. Fable […]

Summer is almost here, and that means getting outside, having fun in the sun, and touching grass. But if you need to cool down for a bit, allow me to sing the praises of the library. If your local library system is anything like mine, it probably offers a lot of great things for you […]

When a platform started with total developer autonomy, teams felt overwhelmed and ended up solving the same problems in completely different ways. The company shifted to enablement over support, working together with teams intensively, and helping teams feel confident and capable, turning the right way into being the easiest way. By Ben Linders

TL;DR - Getting an app to run is now the easy part. AI is very good at producing something that...

It’s the first estimate of how many Americans are sneaking onto Polymarket’s banned crypto-based platform.

Deezer will now help you find AI slop in your music playlists even if you're on another platform.