AI News
Signal over noise. Daily.
一線 AI Lab
11 articlesFrom Hugging Face to Amazon SageMaker Studio in one click
A Blog post by Amazon on Hugging Face
Claude Sonnet 5
Our most agentic Sonnet yet, with top-tier intelligence for coding and everyday professional work.
Hugging Face Models on Foundry Managed Compute
A Blog post by Microsoft on Hugging Face
Agent SDK reference - Python
Complete API reference for the Python Agent SDK, including all functions, types, and classes. pip install claude-agent-sdk The Python SDK provides two ways to interact with Claude Code: query() creates a new session by default and handles a single exchange with connections managed automatically, while ClaudeSDKClient reuses the same session across multiple exchanges with manual connection control. Both support streaming input.
Handle approvals and user input
Surface Claude's approval requests and clarifying questions to users, then return their decisions to the SDK. While working on a task, Claude sometimes needs to check in with users. It might need permission before deleting files, or need to ask which database to use for a new project. Claude requests user input in two situations: when it needs permission to use a tool, and when it has clarifying questions (via the AskUserQuestion tool). Both trigger your canUseTool callback, which pauses execution until you return a response.
Intercept and control agent behavior with hooks
Intercept and customize agent behavior at key execution points with hooks. Hooks are callback functions that run your code in response to agent events, like a tool being called, a session starting, or execution stopping. With hooks, you can block dangerous operations before they execute, log and audit every tool call for compliance, transform inputs and outputs to sanitize data or inject credentials, require human approval for sensitive actions, and track session lifecycle to manage state or send notifications.
Give Claude custom tools
Define custom tools with the Claude Agent SDK's in-process MCP server so Claude can call your functions, hit your APIs, and perform domain-specific operations. Using the SDK's in-process MCP server, you can give Claude access to databases, external APIs, domain-specific logic, or any other capability your application needs. This guide covers how to define tools with input schemas and handlers, bundle them into an MCP server, pass them to query, and control which tools Claude can access. It also covers error handling, tool annotations, and returning non-text content like images.
Quickstart
Get started with the Python or TypeScript Agent SDK to build AI agents that work autonomously. Use the Agent SDK to build an AI agent that reads your code, finds bugs, and fixes them, all without manual intervention. You'll set up a project with the Agent SDK, create a file with some buggy code, and run an agent that finds and fixes the bugs automatically. Requires Node.js 18+ or Python 3.10+ and an Anthropic account. For your own projects, you can run the SDK from any folder; it will have access to files in that directory and its subdirectories by default.
Configure permissions
Control how your agent uses tools with permission modes, hooks, and declarative allow/deny rules. The Claude Agent SDK provides permission controls to manage how Claude uses tools. Use permission modes and rules to define what's allowed automatically, and the canUseTool callback to handle everything else at runtime. When Claude requests a tool, the SDK runs hooks first — a hook can deny the call outright or pass it on. A hook that returns allow does not skip the deny and ask rules; those are evaluated regardless of the hook result.
How the agent loop works
Understand the message lifecycle, tool execution, context window, and architecture that power your SDK agents. The Agent SDK lets you embed Claude Code's autonomous agent loop in your own applications. The SDK is a standalone package that gives you programmatic control over tools, permissions, cost limits, and output. You don't need the Claude Code CLI installed to use it. When you start an agent, the SDK runs the same execution loop that powers Claude Code: Claude evaluates your prompt, calls tools to take action, receives the results, and repeats until the task is complete.
Expanding Managed Agents in Gemini API: background tasks, remote MCP and more
We're announcing new capabilities in Managed Agents in Gemini API so developers can build reliable, production-ready agents.
媒體
30 articlesMeta rolls out Muse, a new AI image generator
The new image-generating model has numerous use cases, including advertising, decorating and creator-based opportunities.
Meta Now Lets Anyone Use Your Instagram Photos in AI Images—Unless You Opt Out
As part of Meta's Muse Image model rollout, Instagram users with public accounts need to opt out to block AI generations of their content.
OpenAI's Chief Futurist Is Leaving the Company
Joshua Achiam spent nearly nine years at OpenAI researching AI safety and made a memorable appearance in the Musk v. Altman trial.
Data centers' energy demand threatens Trump's "Made in America" plan
US manufacturers in many Rust Belt cities and towns are paying significantly higher electricity costs as growing energy demand from data centers strains the largest power grid operator in the United States. The resulting squeeze on profit margins for steelmakers and brick factories could further undermine President Donald Trump's "Made in America" plan to revive US manufacturing, and it comes as Trump has simultaneously championed the tech companies behind the AI data center boom. Factory electricity bills are generally rising faster than those for other business customers or residential customers.
Anthropic Expands in Manhattan, Part of an A.I. Boom in New York
Anthropic, the artificial intelligence company, plans to lease a 16-story building in Hudson Square and to double its work force in New York City to 1,000 people this year.
Meta's new Muse Image model can pull other Instagram users into AI photos
Meta is launching the first AI image generation model made by its Superintelligence Labs division. The Muse Image model now powers the image-making tools across the Meta AI app, Instagram, and WhatsApp, and it's coming soon to Facebook and Messenger. It's part of the growing Muse family of AI models that replace Meta's Llama lineup. Alexandr Wang, who Meta hired to head up its Superintelligence Labs last year, says on Threads that Muse Image is "agentic," meaning it works with its Muse Spark large language model to reason through your prompt, search the web, and plan before it generates.
Why the rise of open source AI isn't hurting Anthropic … yet
Open source models' success isn't coming at the expense of frontier labs. Instead, they each seem to capture two phases of the same life cycle.
Microsoft joins AI cost-cutting trend by relying more on its own models
Microsoft is the latest Silicon Valley giant to cut back on its AI spending.
Meta Unveils an A.I. Image Generator
Muse Image, which can create realistic images for users on Instagram and WhatsApp, is the company's latest attempt to catch up in the global artificial intelligence race.
Discord admits AI moderation bug wrongfully banned users over harmless images
The company confirmed that the issue had been affecting accounts since May, with an additional 200 users banned over the weekend before its team identified and fixed the problem.
Samsung Made More Profit Last Quarter Than the Last Two Years Combined
The A.I. boom has propelled the South Korean tech giant's earnings, but its shares sank as investors had even loftier expectations.
Two A.I. Systems: 'The Frontier and Everything Else'
On the "Hard Fork" podcast, the host Casey Newton acknowledges the gap between frontier A.I. models and other models, including open source, distilled and Chinese models.
Anthropic is launching Claude Cowork on mobile and web
Starting Tuesday, Anthropic's Claude Cowork AI platform will be available on mobile and web for the first time. The expanded access is rolling out first to Max subscribers and coming to Claude users on other plans in the coming weeks. Claude Cowork was previously only accessible through the Claude desktop app for macOS and Windows, but now users on iOS and Android can also use it. However, Anthropic says the full experience for Cowork will still be on the desktop app, including features like local file access. Cowork sessions will also now run in the cloud by default.
What Are Safe A.I. Use Cases for Toddlers?
"Use it to enhance, not replace," advises Dr. Dana Suskind, an early learning specialist who appeared on the "Hard Fork" podcast to share her advice on using A.I. to parent young children.
How Technology Changed the World Cup
The Athletic's Hannah Vanbiber discussed how technology has changed the World Cup with the implementation of tools like 3-D A.I. avatars, goalkeeper line-of-sight cameras, and Connected Ball Technology.
Claude Cowork expands to mobile and web
With this update, users can start a task from their desk, get status updates on their phone, and pick up the finished output later — even if their laptop is closed.
Facing US export controls, China's DeepSeek plans to make its own chips
DeepSeek, the Chinese startup developing large language models that are competitive with those from US companies like OpenAI and Anthropic, is planning to enter the silicon business, according to Reuters. Citing three people familiar with the matter, Reuters writes that DeepSeek has been working on a move into silicon for about a year. It has been meeting with potential partners in the hardware and silicon space and has been hiring engineers for the project.
Shut Those Laptops! Anthropic Puts Its Claude Cowork Agent on Your Phone
Claude Cowork now keeps working on tasks even after you close your laptop. It's part of a larger push toward smartphone-controlled agents.
Solos debuts an even lighter version of its camera-less smart glasses
The AirGo A6 are available in multiple designs including several transparent color options. Solos announced a new version of its AirGo smart glasses, one that forgoes cameras for a sleeker design and an AI assistant that relies on voice interactions. Last year's AirGo A5 weighed 36 to 40 grams depending on the frame style, but the new AirGo A6 weigh around 19 grams. Part of the weight savings comes from thinner temple arms housing speakers, batteries, and other electronics. Pricing and availability for the new AirGo A6 hasn't been finalized yet.
These New Smart Glasses From Solos Come With a Privacy Shield for the Cameras
You can clip a cover over the cameras, which could be a double-edged sword.
Why Index Funds Haven't Boosted SpaceX
Some market watchers are puzzled that shares in Elon Musk's rocket and A.I. company have stalled out despite major interest from Wall Street.
Savi's app aims to protect consumers from realistic AI scams like kidnappers demanding ransom
The company just raised $7 million in seed funding, and is launching its app for iPhone and Android on Tuesday.
The foundational elements of AI architecture that IT leaders need to scale
With the rapid progress of AI capabilities and the move to agentic systems, organizations are expanding their use cases as the technology continues to grow. That constant evolution also introduces risk, leaving IT leaders to wonder which investments will prove valuable even six months into the future. Returning to the foundational elements of AI architecture—the structural framework required for deploying and managing reliable, integrated AI systems at scale—allows technology leaders to make astute decisions today while supporting a future of AI agents that can retrieve information, make decisions, and execute complex workflows across systems.
How AI could enable autonomous robot workers in workplaces—and maybe homes
In a world where self-driving robotaxis glide through major city streets without drivers behind the wheel and delivery drones autonomously fly through the skies to drop off orders at customers' homes, the idea of general-purpose robots helping humans with various tasks in workplaces or even homes may not seem far-fetched. But that future hinges on developing increasingly autonomous robots powered by modern artificial intelligence—an ambitious vision that has motivated many researchers to become startup founders while also attracting billions of dollars in investment.
Erling Haaland Is Everywhere at the World Cup. Most of It Is AI
Norwegian striker Erling Haaland isn't just a footballer anymore. He's become an internet character perpetuated by fans and AI.
British Space Startup Launches Longevity Lab Into Orbit
The lab will beam back data to train AI models to predict how proteins behind age-related diseases like Alzheimer's and certain cancers behave.
Those Soldiers Flooding Your Feeds? They Might Not Be Real.
Videos generated by artificial intelligence have swarmed social media to exploit public support of U.S. troops for other purposes.
The first American autonomous ground vehicles are fighting in Ukraine
Forterra has deployed more than 100 of its self-driving ATVs in conflict zones in Ukraine.
NATO Needs Turkey
The war in Ukraine and Trump's disregard for the alliance have generated a new appreciation of what Turkey brings to the table.
'Who Should I Vote for?' Voters Turn to A.I. Before Casting Their Ballots
It takes effort to be an informed citizen. Artificial intelligence tools offer an alluring shortcut — but they're not without risk.