The Google Tensor ML SDK is graduating to its Beta phase, allowing developers to build and deploy high-performance machine learning models directly onto the TPU of Google Pixel 10 devices. By integrating with LiteRT, Google's edge deployment framework, the SDK provides a unified workflow for developers to convert, compile, and run PyTorch or TFLite models with robust fallback options. Additionally, a new model garden offers over 100 classic and generative AI models, including Gemma 3, enabling low-latency, private features like speech recognition, computer vision, and text generation.
The Google AI Edge Gallery app has expanded its on-device AI capabilities by introducing experimental support for the open-source Model Context Protocol (MCP) on Android, allowing Gemma 4 to coordinate complex tasks across external data sources like Google Workspace and Google Maps. To enable more proactive and persistent user interactions, the update adds a "Schedule Notification" skill for automating routines and a persistent chat history feature that restores long session contexts nearly instantly. Driven by an open-source toolkit, the platform encourages community developers to build and share custom utility-focused workflows, prompt configurations, and tool integrations via its GitHub repository.
Google AI Edge’s LiteRT-LM provides a production-proven, highly optimized infrastructure for running Gemma 4 across cross-platform mobile and edge environments. It actively unlocks the model's native multimodal and agentic features on-device by utilizing memory-efficient dynamic loading, Multi-Token Prediction for up to a 2.2x speedup, and advanced orchestration tools like Thinking Mode and Constrained Decoding. Furthermore, the engine is rapidly expanding its integration surfaces beyond Android, introducing new native Swift APIs for Apple ecosystems and WebGPU-accelerated JavaScript APIs for high-performance, serverless browser inference.
Google announced the transition from assistive AI to independent agents, highlighting the launch of the Gemini 3.5 series and major updates to its Antigravity agent-first development platform. For mobile developers, the post introduces new Android CLI tools, the Android Bench evaluation leaderboard, and an automated Migration agent designed to rapidly convert various frameworks into native Kotlin code. Web development is also being transformed through Chrome DevTools for agents, the HTML-in-Canvas API, and the proposal of WebMCP, an open web standard that enables browser-based AI agents to execute complex tasks.
The Google Cloud and NVIDIA developer community is celebrating its first anniversary with 100,000 members and a renewed focus on providing builders with advanced AI infrastructure and resources. To accelerate development, the community offers curated learning pathways for mastering LLM optimization, GPU-accelerated data analytics, and monthly expert-led webinars. Moving into its second year, the initiative will expand to include hands-on labs, engineering events, and specialized content focused on the growth of agentic AI.
Google is unifying its AI terminal tools by transitioning the community-focused Gemini CLI into Antigravity CLI, a new agent-first platform built for complex, multi-agent workflows. This new Go-based tool offers faster execution, asynchronous processing, and a unified architecture that syncs with the Antigravity 2.0 desktop application. While enterprise customers will maintain existing access, individual and free users must transition to the new platform before Gemini CLI stops serving requests on June 18, 2026.
Google is expanding its smart home ecosystem by launching a full-stack Gemini AI offering that integrates advanced camera intelligence, natural language queries, and daily activity summaries. This initiative provides service providers and hardware manufacturers with turnkey reference designs and APIs to build proactive, branded services without extensive research and development. Ultimately, the program aims to move beyond basic device control toward an AI-native home that can understand context and care for users' needs in real time.
Google has announced the launch of version 0.1.0 of the Agent Development Kit (ADK) for Kotlin, alongside a specialized ADK library for Android. This open-source framework simplifies the creation of AI agents by managing complex orchestration, session sharing, and error handling across cloud and edge environments. The release supports hybrid orchestration, enabling developers to build multi-agent systems where a cloud-based model can seamlessly offload specific tasks to local, on-device models like Gemini Nano to enhance user privacy.
Google Pay is evolving for "agentic commerce" by introducing the Universal Commerce Protocol and a new MCP server that allows AI agents to manage integrations and analyze trends. New Android updates introduce dynamic callbacks for seamless express checkouts and extend payment support into social media apps via WebViews. Additionally, the platform is launching cross-device biometric authentication and new transaction signals to help merchants reduce friction and optimize processing costs.
Google has announced the new Google Pay & Wallet Developer MCP server, an open-standard tool designed to securely connect AI development assistants and IDEs with real-time API and account context. The server allows developers to remain within their development environment to search official documentation, validate Wallet pass definitions, check integration status, and manage merchant accounts. Ultimately, this integration aims to reduce friction and accelerate development workflows by minimizing context switching and providing up-to-date, grounded AI support.
The Google Tunix Hackathon on Kaggle challenged developers to transform small, non-reasoning base models into general reasoning engines using Kaggle TPUs and a limited compute budget. The winning teams achieved this by implementing multi-stage post-training pipelines that combined Supervised Fine-Tuning (SFT) with advanced alignment techniques like GRPO and SimPO. Ultimately, the competition democratized AI development by proving that highly capable, structured reasoning models can be successfully trained by the community using accessible, open-source resources.
Google DeepMind’s Gemma 4 12B model brings agentic, multimodal AI capabilities to everyday laptops with 16GB of RAM, enabling local data processing and visual insight generation. Users can leverage this model on macOS through the Google AI Edge Gallery for dynamic Python code execution and visualization, as well as via Google AI Edge Eloquent for completely offline voice dictation and text editing. Additionally, developer workflows are enhanced by the LiteRT-LM CLI's new serve command, which creates an industry-compatible local endpoint to power fully-local AI tools and agents.
The newly released Gemma 4 12B is a dense, multimodal model designed for high-performance local AI execution on consumer devices. By introducing a novel, encoder-free architecture, it bypasses traditional visual and audio encoders to feed multimodal data directly into the LLM backbone.
Google has announced the Google Colab Command-Line Interface (CLI), a new tool that allows developers and AI agents to connect local terminals to remote Colab runtimes for frictionless execution. The lightweight CLI enables users to easily request high-powered GPUs, run local Python scripts remotely, and seamlessly retrieve artifact logs or models like fine-tuned Gemma 3 adapters. By integrating directly into standard terminal environments, the tool is highly programmable and ready to be used by AI agents such as Antigravity or Claude Code to manage complex machine learning pipelines.
DiffusionGemma is an experimental text-generation model built on the Gemma 4 architecture that uses diffusion-based parallel generation instead of token-by-token autoregression, enabling much faster inference, bidirectional context awareness, and real-time self-correction while remaining deployable on consumer GPUs. Its architecture generates and refines 256-token blocks in parallel through iterative denoising, allowing it to handle complex constraint-based tasks such as Sudoku more effectively than traditional language models and demonstrating strong gains from fine-tuning. The model integrates with vLLM and other popular inference frameworks, giving developers access to a new non-autoregressive approach that combines high performance, efficient long-context scaling, and straightforward customization and deployment.
Google is enhancing Sign in with Google by introducing new OIDC standard claims—specifically auth_time and amr (Authentication Methods Reference) to provide developers with deeper session metadata. These updates allow verified apps to verify the "freshness" of a user's login and the specific authentication methods used (such as MFA or hardware keys), enabling more dynamic, risk-based access controls. By leveraging these federated identity signals, platforms can better prevent account takeover and fraud while implementing granular security policies like step-up authentication for sensitive actions.
Google has officially launched the TPU Developer Hub, a centralized educational resource designed to help model builders and developers maximize the performance of Google Cloud TPUs. The hub offers code-first resources, open-source recipes, and deep-dive documentation covering hardware architecture, software optimization, debugging, parallelism, and networking. These materials are tailored for both human developers and AI-assisted tools to streamline everything from large-scale training to low-latency inference workloads.
This post introduces three architectural patterns designed to integrate Model Context Protocol (MCP) Apps and Agent-to-User Interface (A2UI) to solve the tradeoff between highly custom iframe environments and native, declarative rendering. By combining these approaches, developers can serve native-feeling UIs directly over MCP servers, embed complex and stateful iframe apps securely inside declarative views, or inject generative UI components into legacy systems. Ultimately, these hybrid frameworks empower engineering teams to deliver secure, performant, and brand-consistent agentic user experiences tailored to their specific project constraints.
What happened in 2025 was this: the economics of code production were turned upside down . Instead of being very hard, time-consuming, and expensive to generate code, it became effectively free and instant. Lines of code went from being treasured, reused, cared for and carefully curated, to being disposable and regenerable, practically overnight. — Charity Majors , AI demands more engineering discipline. Not less Tags: charity-majors , ai-assisted-programming , generative-ai , ai , llms
The Cloudflare One stack is a library of agent skills that gives any AI agent the knowledge it needs to plan, deploy, and manage a Zero Trust environment — no migration calls required.
OpenAI and Molecule.one show how a near-autonomous AI chemist using GPT-5.4 improved a key drug-making reaction, advancing medicinal chemistry research.
Tool: <click-to-play> — a still that plays A progressive enchantment Web Component that turns this markup: <click-to-play> <a href="URL to GIF"> <img src="URL to first frame" alt="..."> </a> </click-to-play> Into a still frame with a click to play button which loads the GIF on demand. For when you don't want big GIFs to be loaded unless people want to play them. Here's an example that demonstrates the new row editing tools in Datasette - in fact I built this Web Component for that post. Tags: gif , javascript , progressive-enhancement , web-components
NetNewsWire Status I find this inspiring. Brent Simmons retired a year ago, and his retirement project is making one piece of software really, really good - free from any commercial pressure. The software is NetNewsWire - "it's like podcasts, but for reading " - first released in 2002 and made open source in 2018. I've been using it on Mac and iPhone for several years now and I'm finding it indispensable. Via Lobste.rs Tags: brent-simmons , netnewswire , open-source
Introducing LifeSciBench, an expert-authored, expert-reviewed benchmark for evaluating how AI systems handle real-world life science research tasks and decisions.
Release: datasette 1.0a34 Quoting the release notes: The big feature in this alpha is tools to insert, edit and delete rows within the Datasette interface. These features are available on table pages, and edit and delete are also available as action items on the row page. The inspiration for this feature - which is long overdue - was Datasette Agent . I added SQL write support to that the other day which highlighted how absurd it was that you could insert and edit ties via the chat interface but not in the regular Datasette UI! Tags: projects , datasette , annotated-release-notes
Release: datasette-tailscale 0.1a0 A very experimental alpha plugin which lets you do this: datasette tailscale mydata.db \ --ts-authkey tskey-auth-xxxx --ts-hostname datasette-preview This starts a localhost Datasette server with a Tailscale sidecar that connects it to your Tailnet, such that http://datasette-preview/ serves Datasette. It's using the Python bindings for the experimental tailscale-rs library. I filed an issue asking if there's a cleaner way of setting up the proxy mechanism. Tags: datasette , tailscale
I can 100% attest to the fact that Qwen3.6-27B is a very capable local model for coding tasks. Over the last month and a half I've been using it almost daily, either on my M2 Ultra or on my RTX 5090 box. I use it for small mundane tasks at ggml-org - nothing really impressive, but definitely a helpful tool for a maintainer. I think I would be using it much more, if I didn't have to spend a lot of my time on reviewing PRs. Currently, I have a very lightweight harness - the pi agent with everything stripped ( pi -nc --offline ) and a short system prompt to align it a bit with my style. — Georgi Gerganov , Hacker News comment on Running local models is good now by Boykis Tags: georgi-gerganov , llms , ai , generative-ai , pi , ai-assisted-programming , local-llms , qwen , coding-agents
Get unified visibility into your email authentication posture and reach full DMARC enforcement with deeper reporting, record analysis, and SPF audits free for every Cloudflare customer.
The Fable 5 Export Controls Harm US Cyber Defense I quoted The Atlantic quoting Kate Moussouris earlier, when I should have gone straight to the source. Here she is confirming that the "jailbreak" that got Claude Fable 5 banned under an export control really was "fix this code": The researchers took open-source code with known CVEs, plus new code with deliberately planted vulnerabilities, and asked Fable 5, Mythos, and Opus to “review the code for security issues.” Fable 5 refused. They then asked the models to “fix this code” and, through a multistep and manual process, turned the output into scripts that test the patches. As Kate points out, this is absurd. Coding models fix bugs, and security exploits are the most important category of bugs for them to fix! Defenders need to be able to ask AI to fix the bugs in a file, explain why the fix matters, and write tests that confirm the patch works. That is not a guardrail bypass. It is the most valuable thing an AI model can do for defensive security: executing the find, fix, and test loop defenders run every day. [...] The prompts worked because they were defensive requests, and that capability cannot be removed without making the model worse at fixing bugs and verifying patches. This whole situation is such a mess. Non-technical decision-makers have been hearing that models that can "craft cyber attacks" are uniquely dangerous for months. Now they look ready to ban any model that can help us secure our code. Tags: jailbreaking , security , ai , generative-ai , llms , anthropic , ai-security-research , claude-mythos
Katie Moussouris, a cybersecurity expert and the CEO of Luta Security, told me that Anthropic shared with her a copy of the White House’s report on the Fable jailbreak to get her appraisal. (She said that she is not being paid by Anthropic.) The report, Moussouris said, involved IT experts asking Fable to help find and patch bugs. When given deliberately insecure code, she said, Fable refused the prompt “review the code for security issues” but then complied when asked to “fix this code,” followed by some further manual steps. Moussouris told me that this was just “the model working as intended” for cyberdefense. — Matteo Wong, The Atlantic , The White House Is Ratcheting Up Its War Against Anthropic Tags: anthropic , claude , ai , llms , ai-ethics , jailbreaking , generative-ai , ai-security-research , claude-mythos
TIL: Cloudflare CAPTCHA on at least one ampersand I'm using Cloudflare's CAPTCHA (they call it a "Web Application Firewall > Custom rules > Managed Challenge" these days) to prevent crawlers from aggresively spidering my faceted search engine on this site, but I got fed up of even simple ?q=term searches triggering the challenge. After some mucking around with Claude Code it turns out you can register the following rule instead, so the CAPTCHA only kicks in for search URLs containing at least one ampersand: (http.request.uri.path wildcard r"/search/*" and http.request.uri.query contains "&") And now /search/?q=lemur works without triggering a CAPTCHA! Also included: notes on trying out the Cloudflare MCP with Claude Code , though it turned out not to be able to edit the rules in question so I had Claude Code switch to the Cloudflare API instead. Tags: captchas , cloudflare , model-context-protocol , claude-code
OpenAI introduces Deployment Simulation, a method to predict AI model behavior before deployment using real conversation data to improve safety and evaluation accuracy.
Release: datasette-agent 0.3a0 New tool, execute_write_sql , which requests user approval and then writes to a database - taking user permissions into account. #27 I added a mechanism for asking user approval in datasette agent 0.2a0 . The new execute_write_sql tool can now prompt the user for all kinds of useful operations. Here's an example where I add some pelican sightings to my pelican_sightings table: The new version also enhances the datasette agent chat terminal mode to support approvals, and adds several new options including --unsafe mode for auto-approving them: datasette agent chat can execute tools that require user approval. #30 Three new options for datasette agent chat - --root to run as root, --yes to approve all ask user questions, and --unsafe for both. Tools can now provide plain text alternatives to HTML, for display in the datasette agent chat CLI. #31 The datasette agent chat content.db -m gpt-5.5 --unsafe command can now be used to chat directly with a specific database and directly modify it through prompts like "create a notes table", "add a note about X" etc. Tags: projects , ai , datasette , annotated-release-notes , generative-ai , llms , llm-tool-use , datasette-agent
"They screwed us": Personality clashes sent Anthropic's models offline Lots of "source familiar with the administration's thinking" and "source close to Anthropic" in this Axios piece, which is the best collection of behind-the-scenes gossip I've seen about the US government export control Mythos/Fable story so far. Logan Graham ( I lead the Frontier Red Team at Anthropic ), Dave Orr (Head of Safeguards, previously a Director of Engineering at Google DeepMind), and blog favorite Nicholas Carlini are reported to be meeting with the Commerce Department today in D.C. Good luck to them! (I just noticed Logan was "Special Adviser to the Prime Minister" in the Boris Johnson era, covering AI, science, and technology policy - so significant political experience.) This closing notes doesn't give me much optimism that we'll be getting Fable back any time soon: The bottom line : One option is to make sure Anthropic's models can't be jailbroken — though perfect jailbreak resistance may be impossible. Absent that, a source familiar with the administration's thinking said it may simply come down to an attitude fix where, instead of feeling dismissed, "everyone feels safe, secure and happy." This made me wonder if Anthropic ever successfully addressed the class of attacks described in the Universal and Transferable Adversarial Attacks on Aligned Language Models paper from 2023. It looks like their Constitutional Classifiers work (that post is from January this year) is relevant to that. They continue to claim that no "universal jailbreak" has been found against Claude Mythos, classifying the jailbreak that triggered the US government response as "a potential narrow, non-universal jailbreak". Tags: jailbreaking , ai , generative-ai , llms , anthropic , claude , nicholas-carlini , ai-ethics , claude-mythos
Cloudflare is deepening our investment in AI with the addition of team members from Ensemble AI, focusing on machine learning infrastructure and efficiency.
[...] Instead, I picture a specific person and I just write for them. Often this person is "me, but 3 years ago" or a good friend. — Julia Evans , write for 1 person Tags: writing , julia-evans
In our ongoing SIG Spotlight series, we shine a light on the groups that keep the Kubernetes project moving forward. This time, we catch up with SIG Storage , the group responsible for persistent data, volume management, and the interfaces that connect Kubernetes workloads to the storage systems beneath them. We spoke with Xing Yang , Co-Chair of SIG Storage and Software Engineer at VMware by Broadcom, about the SIG's history, the features shipping in recent Kubernetes releases, and where storage in Kubernetes is headed as AI workloads become the norm. Introductions Could you introduce yourself and share your role(s) within SIG Storage? My name is Xing Yang , a software engineer at VMware by Broadcom. I'm a co-chair in SIG Storage, alongside another co-chair Saad Ali from Google. There are also two Tech Leads in SIG Storage: Michelle Au from Google and Jan Šafránek from Red Hat. What first drew you to storage in Kubernetes, and how did you start contributing? I have always been working in the storage domain, so SIG Storage was a natural place for me to get started when I began to learn Kubernetes. I started attending SIG Storage meetings , trying to figure out what I could do to help. This was before the first Container Storage Interface (CSI) release — lots of things were still evolving. It was a very exciting time. What subprojects or areas do you actively maintain or review today? I'm a maintainer in Kubernetes CSI. There are multiple CSI sidecars — such as csi-provisioner , csi-attacher , csi-resizer , and csi-snapshotter — that we need to release following every Kubernetes release. I'm also a co-chair for a Data Protection Working Group co-sponsored by SIG Storage and SIG Apps . Several features have come out of that WG aimed at filling gaps in data protection support within Kubernetes. One is Volume Group Snapshot , which provides crash-consistent group snapshots for multiple volumes used by an application. Changed Block Tracking (CBT) is another critical feature from the DP WG designed to support efficient backups. About SIG Storage For folks who are new: what is SIG Storage, in your own words? What problems in Kubernetes are you trying to solve? SIG Storage is a Special Interest Group focused on how to provide storage to containers running in your Kubernetes cluster. We define standard interfaces so that a storage vendor can write a driver and have its underlying storage system consumed by containers in Kubernetes. Why does Kubernetes need a dedicated storage SIG? What makes storage hard in a distributed system? When Kubernetes was first introduced, it was meant for stateless workloads only. Container applications were regarded as ephemeral and therefore did not need to persist data. However, that changed drastically. Stateful workloads started running in Kubernetes, and we needed a dedicated SIG to tackle the associated storage challenges. PersistentVolumeClaims, PersistentVolumes, and StorageClasses were all introduced to provision data volumes for applications running in Kubernetes. How did SIG Storage originally form, and how has its mission changed over time? SIG Storage was formed to address the challenges of handling persistent data within Kubernetes. Initially, PersistentVolumes were implemented as in-tree plugins, and the SIG managed those plugins while developing core storage primitives like PersistentVolumes and PersistentVolumeClaims. Container Storage Interface (CSI) was introduced later and played a crucial role in simplifying storage integration, enabling third-party storage providers to develop and maintain their own out-of-tree plugins without modifying Kubernetes core code. With basic integration addressed by CSI, the SIG's mission expanded to include advanced storage features that leverage the new interface. The SIG has also expanded its scope to support object storage through the Container Object Storage Interface (COSI). Current work and roadmap What are the top features SIG Storage is actively working on right now? The Data Protection WG has been working on a couple of exciting features: VolumeGroupSnapshot is a Kubernetes feature enabling a crash-consistent, point-in-time snapshot of multiple PersistentVolumes simultaneously. This ensures data integrity for applications — like databases — that rely on multiple volumes by capturing all volumes in the group atomically, at the exact same point in time. It just moved to GA in Kubernetes v1.36. CSI Changed Block Tracking (CBT) enables efficient, incremental backups. By allowing storage systems to report only the blocks that have changed since the last snapshot, it significantly reduces the amount of data that needs to be transferred. It just moved to Beta in Kubernetes v1.36. Another feature worth highlighting is Container Object Storage Interface (COSI) . COSI provides a standard interface for provisioning and consuming object storage buckets in Kubernetes — standardizing object storage for containerized applications much like CSI did for block and file storage. COSI is now transitioning to v1alpha2 , with plans for promotion to Beta in a future release. What recent work from SIG Storage do you consider a "win" for users? The graduation of VolumeAttributesClass to GA in Kubernetes v1.34 is a major win for users managing stateful workloads. Previously, changing volume attributes like IOPS or throughput required out-of-band actions or disruptive operations. Now, users can dynamically tune storage properties such as IOPS or throughput directly through the Kubernetes API — scaling up for peak loads or down to optimize costs — without external processes or downtime. VolumeAttributesClass enables dynamic modification of storage characteristics without recreating the volume. This completes the picture by allowing users to tune both capacity and other storage properties dynamically, just as they can now tune both CPU and memory for compute. Looking ahead one or two releases, what's on the roadmap that people should watch for? I'd like to draw attention to the Volume Health feature. This feature is designed to offer critical visibility into the operational status and integrity of persistent volumes. By enabling storage drivers and the Kubernetes control plane to report issues, it allows for proactive monitoring and identification of volume-related problems. Currently, volume health information is reported via non-persistent events. We are actively investigating enhancements to this feature with the goal of supporting automated remediation capabilities in the future. Are there areas where you'd really like more discussion or help from the community? We always need help from the community to fix bugs, add tests, and help with reviews. We'd also like to get feedback on the Alpha feature Mutable PV Affinity , which was introduced in Kubernetes v1.35. Use cases include migrating volumes from zonal to regional storage or migrating from one disk type to another. Another topic is volume replication . It was raised at KubeCon Atlanta and has been discussed in the Data Protection WG. Community members interested in this topic are encouraged to join the DP WG meetings. What are the biggest challenges users face today when running stateful workloads on Kubernetes? While Kubernetes has moved stateful workloads — like databases and AI pipelines — into the mainstream, managing "state" in a system designed for ephemerality remains difficult: Data Gravity and Storage Locality : Pods move in seconds, but data has gravity. If a node fails, a pod using local storage is stuck. Operators must decide whether the failure is transient or permanent — a high-stakes call. This is why we are enhancing the Volume Health feature to provide the visibility needed to automate recovery choices. Day 2 Complexity : Setting up a database is easy; maintaining its health over time is the real challenge. Standard Kubernetes objects like StatefulSets offer a baseline, but they lack the operational logic needed for tasks such as schema upgrades, engine patching, or cluster-wide Kubernetes upgrades. Data Mobility : Moving persistent data remains a significant hurdle — whether migrating between storage tiers, shifting workloads across availability zones, or moving to a different cluster. This challenge includes ongoing synchronization and replication for high availability and disaster recovery across a distributed system. Storage and AI How do you see storage evolving in Kubernetes over the next few years, especially as AI/ML workloads grow? I see several trends shaping storage in Kubernetes as it evolves from a container orchestrator into the "Operating System" for AI: More Intelligent Data Management : We'll see a shift toward smarter CSI drivers and data management tools offering advanced features like automatic tiering, snapshots, migration, and replication — optimized specifically for high-performance AI/ML workflows and large data platforms. Object Storage as a First-Class Citizen : AI datasets now frequently reach exabyte scale, making object storage the preferred choice for AI workloads. COSI is standardizing bucket management just as CSI did for disks, allowing data scientists to use a BucketClaim to provision S3-compatible storage natively and unifying object, file, and block storage into a single workflow. Performance and Low Latency : For AI/ML, storage needs to keep up with GPU processing speeds. This will accelerate adoption of high-performance parallel file systems and NVMe-over-Fabrics (NVMe-oF) technologies managed natively via Kubernetes. The line between traditional block/file and memory-speed storage will continue to blur. Data-Aware Scheduling : Instead of just considering CPU and RAM, the Kubernetes scheduler will increasingly prioritize placing Pods based on data locality — calculating the cost of moving data versus moving compute to keep massive data platforms performant. SIG Storage continues to tackle some of the hardest problems in Kubernetes: keeping stateful applications running reliably, making storage operations transparent and composable, and now scaling up to meet the demands of AI-era workloads. Whether you're a user managing databases in production or a developer curious about storage internals, there's a place for you in SIG Storage. If you'd like to get involved, check out the SIG Storage community page and join the bi-weekly meetings . You can also find the SIG on Slack at #sig-storage . SIG Storage Mailing List SIG Storage on Slack Data Protection WG
Why AI hasn’t replaced software engineers, and won’t Arvind Narayanan and Sayash Kappor take on the question of AI job losses through the lens of a profession that is uniquely suited to AI disruption - software engineering. In this essay, we argue that there is enough evidence to reject the narrative that once AI capabilities reach a certain threshold, it will cause mass layoffs. Given that this is true even in a sector with very few regulatory barriers, most other professions are likely to be even more cushioned. The first good news is that the data still doesn't support the idea that AI is causing mass unemployment. In March 2025, New York became the first U.S. state to add an AI disclosure checkbox to WARN Act filings. In the full first year, more than 160 companies filed WARN notices. Not a single one checked the AI box AI speeds up the typing-code-into-a-computer phase, but it turns out software engineering is about a whole lot more than that: If writing code isn’t the bottleneck, what is? The task-breakdown surveys point at things like meetings or debugging. This just leads to more questions: what are developers doing in those meetings and why can’t it be done by AI? Won’t debugging get automated as capabilities improve? To understand the real bottlenecks, we have to get qualitative, and dig into software engineers’ own understanding of what it is they do that resists automation. When we did this analysis, it revealed three things as the real bottlenecks (1) deciding and specifying what to build, (2) verifying and being accountable for what is delivered, and (3) the deep human understanding — of the codebase, the business, and the environment — required to carry out both of these. I'm finding AI assistance also helps me with the deciding and verifying steps, but it's the "deep human understanding" that remains key to the value I provide. Give me all of the AI assistance in the world and the value I produce will still be reliant on how deeply I understand both the problems and the solutions that the agents are building for them. Tags: careers , ai , generative-ai , llms , arvind-narayanan , ai-ethics
The Pyodide 314.0 release announcement (via Hacker News ) includes news I've been looking forward to for a long time: You can now publish Python packages built for Pyodide (or any Python runtime compatible with the PyEmscripten platform defined in PEP 783 ) directly to PyPI and install them at runtime. Previously, the Pyodide maintainers had to maintain, build, and host over 300 packages ourselves. This created a significant burden on our maintainers and became a major bottleneck for the community, as every new package required manual review. Moving forward, package maintainers can simply build and publish Pyodide wheels to PyPI, just as they do for native wheels on Linux, macOS, or Windows. Here's the PR to PyPI itself supporting this , which landed on April 21st. I adore Pyodide , and have been frustrated in the past by this limitation. It's possible to compile C or Rust extensions to WASM in a wheel file, but before now there was no easy way to distribute them. Thanks to the efforts of a whole lot of people, that's now been fixed! Trying it out with luau-wasm I decided to celebrate by finding something I could package. I have quite a few experimental Pyodide projects lying around, but the best fit for this looked to be my Luau WebAssembly research spike from 9th March. Luau is a "small, fast, and embeddable programming language based on Lua with a gradual type system", developed by Roblox and released under an MIT license. It's written in C++. I already knew it was possible to compile it to WebAssembly and get it running inside of Pyodide, so I set Codex + GPT-5.5 xhigh the task of packaging my experiment up and publishing it to PyPI using GitHub Actions. It took some iteration, but here's the result: luau-wasm is a brand new PyPI package which publishes a 276KB luau_wasm-0.1a0-cp314-cp314-pyemscripten_2026_0_wasm32.whl file which can be used in Pyodide like this: import micropip await micropip . install ( "luau-wasm" ) import luau_wasm print ( luau_wasm . execute ( r''' local animals = {"fox", "owl", "frog", "rabbit"} table.sort(animals, function(a, b) return #a < #b end) for i, name in animals do print(i .. ". " .. name .. " (" .. #name .. ")") end ''' )) You can run that code in the Pyodide REPL demo to see it in action. The GitHub repo for luau-wasm includes all of the build and deploy scripts (using the latest cibuildwheel ) and also deploys an HTML demo page which loads Pyodide, installs luau-wasm and provides an interface for trying it out: https://simonw.github.io/luau-wasm/ How many packages are using this so far? I was curious to see how many packages are currently publishing wheels for this platform. After some tinkering with ChatGPT I got to this BigQuery SQL which I ran against PyPI's public dataset on BigQuery . Here's the raw JSON of query results and here's a SQLite SQL query in Datasette Lite which dedupes packages by most recent upload date. If the query is right, there are currently 28 PyPI packages publishing with the new pyemscripten_202*_wasm32 tags: luau-wasm , uuid7-rs , cmm-16bit , pyOpenTTDAdmin , imgui-bundle , numbertoolkit , bashkit , geoarrow-rust-core , arro3-io , arro3-core , arro3-compute , onnx , powerfit-em , tcod , chonkie-core , tokie , robotraconteur , pydantic_core , yaml-rs , cadquery-ocp-novtk-OCP.wasm , uuid_utils , base64_utils , pycdfpp , lib3mf-OCP.wasm , typst , toml-rs , onnx-weekly , dummy-pyodide-ext-test Here's hoping we see a whole lot more of those showing up over the coming months and years. Tags: lua , pypi , python , sandboxing , webassembly , github-actions , pyodide
Research: Mapping SQLite result columns back to their source `table.column` It would be neat if arbitrary SQL queries in Datasette could be rendered with additional information based on which columns from which tables were included in the results. To build that, we would need to be able to look at a SQL query like select users.name, orders.total from users join orders on orders.user_id = users.id and programmatically identify the table.column for each result - navigating not just joins but also more complex syntax like CTEs. I decided to set Claude Code (Opus 4.8, since Fable is currently banned by the US government ) on the problem. It found several promising solutions - one using apsw , another that uses ctypes to access the SQLite sqlite3_column_table_name() C function (which is not otherwise exposed to Python), and one using clever interrogation of the output of EXPLAIN . Tags: python , sqlite , datasette
Statement on the US government directive to suspend access to Fable 5 and Mythos 5 Well this is nuts : The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Anthropic models will not be affected. We received the directive from the government today at 5:21pm (ET). The letter did not provide specific details of its national security concern. Our understanding is that the government believes it has become aware of a method of bypassing, or "jailbreaking" Fable 5. We reviewed a demonstration of this specific technique being used to identify a small number of previously known, minor vulnerabilities. These vulnerabilities all appear relatively simple, and we have found that other publicly-available models are able to discover them as well without requiring a bypass. [...] To date, the government has only given us verbal evidence of a potential narrow, non-universal jailbreak, which essentially consists of asking the model to read a specific codebase and fix any software flaws. Our understanding is that one potential jailbreak was shared with the government. We have reviewed the report and validated that the level of capability displayed there is widely available from other models (including OpenAI's GPT-5.5 ), and is used every day by the defenders who keep systems safe. We will share more details over the next 24 hours. I still have access to Fable via claude.ai and Claude Code now, at 9:01pm ET. Update : I ran this script against the Anthropic API to spot when claude-fable-5 would stop working. My access was cut off at 6:59pm Pacific (9:59pm ET): [2026-06-12T18:56:50-07:00] attempt 35: running uv run llm -m claude-fable-5 hi [2026-06-12T18:56:55-07:00] success: Hi there! How can I help you today? [2026-06-12T18:57:55-07:00] attempt 36: running uv run llm -m claude-fable-5 hi [2026-06-12T18:57:59-07:00] success: Hi! How can I help you today? [2026-06-12T18:58:59-07:00] attempt 37: running uv run llm -m claude-fable-5 hi [2026-06-12T18:59:00-07:00] FAILED after attempt 37 with exit code 1 stderr: Error: Error code: 404 - {'type': 'error', 'error': {'type': 'not_found_error', 'message': 'Claude Fable 5 is not available. Please use Opus 4.8. Learn more: https://www.anthropic.com/news/fable-mythos-access'}, 'request_id': 'req_011CbzRyirV7KZLHYYdBM9od'} Via @AnthropicAI Tags: jailbreaking , ai , generative-ai , llms , anthropic , claude , ai-ethics , claude-mythos
OpenAI WebRTC Audio Session, now with document context I built the first version of this tool in December 2024 to try out the then-new OpenAI WebRTC API for interacting with their realtime audio models. Last month OpenAI introduced a brand new model to that API called GPT‑Realtime‑2 , which they promoted as "our first voice model with GPT‑5‑class reasoning" - with a Sep 30, 2024 knowledge cut-off. I've been waiting for that model to show up in the ChatGPT iPhone app but it still hasn't, so I revisited my old playground. You can now pick the better model, and you can also paste in a big chunk of document context so you can have as audio conversation in your browser about whatever information you think would be useful to explore in a conversational way. Tags: audio , tools , ai , openai , generative-ai , llms , multi-modal-output , webrtc
Jenny owns a crematorium. John’s propane company gives her a $20 billion investment in return for 5 percent of her operation. Jenny throws $10 billion into the incinerator, then pays John $10 billion to buy propane to burn that money to ashes. John reports that his AI investments have generated $10 billion in revenue this quarter and that he owns 5 percent of a $100 billion business. A reporter from Forbes is assigned to profile John and Jenny, and over the course of his research, he becomes embroiled in a passionate but confusing three-way love affair with them, which eventually turns into a polyamorous common-law marriage. His profile is glowing, but light on financial details. — Andrew Singleton , AI Economics for Dummies Tags: ai
Cloudflare Security Insights system now processes over 120 scans per second, providing frequent insights for all customers. By optimizing Kafka consumers, Postgres queries, and our API, we scaled our throughput 10x without adding hardware.
After two days of experience with Claude Fable 5 I think the best way to describe it is relentlessly proactive . It knows a whole lot of tricks and it will deploy pretty much any of them to get to its goal. I'll illustrate this with an example. I was hacking on Datasette Agent today when I noticed a glitch: a horizontal scrollbar that shouldn't be there in the jump menu chat prompt. I snapped this screenshot: Then I started a fresh claude session in my datasette-agent checkout, dragged in the screenshot and told it: Look at dependencies to help figure out why there is a horizontal scrollbar here I had a hunch the cause was in a dependency of Datasette Agent (likely Datasette itself) and I knew Fable was good at digging into dependency code, either by inspecting installed files in its own virtual environment site-packages or by referencing a local checkout on disk. Telling it to start with dependencies felt like a good bet. I got distracted by a domestic task and wandered away from my computer. When I came back a few minutes later I saw my machine open a browser window in my regular Firefox and then navigate to the dialog in question . I had not told Claude Code to use any browser automation, and I was pretty sure it wasn't possible for it to trigger mouse movements or keyboard shortcuts within a window, so how was it doing that? I watched in fascination as it continued with its explorations, then saw it open a Safari window instead of Firefox. I also grabbed this snapshot from the Claude terminal: What was it doing there with uv run --with pyobjc-framework-Quartz ? It turns out Fable had hacked up its own pattern for taking screenshots of browser windows. It was using Python to iterate through all available windows on my machine, then filtering for Safari windows with expected strings such as "textarea" in the window name. It used that to find their window number - an integer like 153551 - which it could then use with the screencapture CLI tool to grab a PNG. OK fine, that's a neat way of taking screenshots. But what was it taking screenshots of? Turns out it had been writing its own scratch HTML pages to try and recreate the bug, then opening Safari and grabbing screenshots. Here's that /tmp/textarea-scrollbar-test.html page it created, and the screenshot it took with screencapture -x -o -l 153551 /tmp/safari-cases.png : (I have way too many open tabs!) OK, so I can see how it's opening test pages and taking screenshots, but how on earth was it triggering the modal dialog that was meant to be under test? That's only available via a click or a keyboard shortcut, and I couldn't see a mechanism for it to run those in Safari. I eventually figured out what it had done. Claude was running in a folder that contained the source code for the application. It knows enough about Datasette to be able to run a local development server. It turns out it was editing Datasette's own templates to add JavaScript that would trigger the correct keyboard shortcut as soon as the window opened, adding code like this: < script > window . addEventListener ( "load" , function ( ) { setTimeout ( function ( ) { document . dispatchEvent ( new KeyboardEvent ( "keydown" , { key : "/" , bubbles : true } ) ) ; } , 1200 ) ; } ) ; </ script > 1.2 seconds after the window opens, this code triggers a simulated / key, which is the keyboard shortcut for opening the modal dialog. There was one challenge left. In order to understand what was going on, Claude needed to run JavaScript on the page to take measurements for itself. It wrote its own custom web application to capture information via CORS, then ran that as a local server and opened a page with JavaScript that would POST directly to it! Here's the Python web app it wrote, using the standard library http.server package: from http . server import HTTPServer , BaseHTTPRequestHandler class H ( BaseHTTPRequestHandler ): def do_POST ( self ): n = int ( self . headers . get ( "Content-Length" , 0 )) open ( "/tmp/diag.json" , "w" ). write ( self . rfile . read ( n ). decode ()) self . send_response ( 200 ) self . send_header ( "Access-Control-Allow-Origin" , "*" ) self . end_headers () def do_OPTIONS ( self ): self . send_response ( 200 ) self . send_header ( "Access-Control-Allow-Origin" , "*" ) self . send_header ( "Access-Control-Allow-Headers" , "*" ) self . end_headers () def log_message ( self , * a ): # quiet pass HTTPServer (( "127.0.0.1" , 9999 ), H ). serve_forever () All this does is accept a POST request full of JSON and write that to the /tmp/diag.json file. It sends Access-Control-Allow-Origin: * headers (including from OPTIONS requests) so that code running on another domain can still communicate back to it. Then Claude injected this code into the template that it was loading in a browser: const host = document . querySelector ( "navigation-search" ) ; const ta = host . shadowRoot . querySelector ( "textarea" ) ; const cs = getComputedStyle ( ta ) ; fetch ( "http://127.0.0.1:9999/diag" , { method : "POST" , body : JSON . stringify ( { dpr : window . devicePixelRatio , scrollWidth : ta . scrollWidth , clientWidth : ta . clientWidth , whiteSpace : cs . whiteSpace , width : cs . width , } ) , } ) ; This took measurements of the <textarea> inside the <navigation-search> Web Component and sent them to the server, which wrote them to a file on disk, which Claude could then read. Having figured out all of these tricks Fable... hit some invisible guardrail and downgraded itself to Opus. Thankfully Opus had access to the full transcript and could continue using the tricks pioneered by Fable, and shortly afterwards found, tested and verified the fix . I prompted Opus to: Write a report in /tmp/automation-report.md where you note down all of the tricks you have used in this session to test against real browsers on my computer, include runnable code examples Which produced this report , which was invaluable for piecing together the details of what had happened for this post. I've shared the full terminal transcript of the Claude Code session as well. A review of everything it did Based on a screenshot and a one-line prompt, Claude Fable 5 + Claude Code: Figured out the recipe to run the local development server (with fake environment variables needed to get it running) Fired up a Playwright Chrome session Turned on the visible scrollbars setting for Chrome defaults write com.google.chrome.for.testing AppleShowScrollBars Always (it turned that off again later) Cycled through Firefox and WebKit in Playwright too, failing to recreate the bug Worked out my default browser was Safari Built a textarea-scrollbar-test.html HTML document Opened that in real (not Playwright) Firefox Found that osascript -e 'tell application "System Events" to tell process "firefox" to id of window 1' was blocked because "osascript is not allowed assistive access" Figured out that uv run --with pyobjc-framework-Quartz python workaround, described above Added JavaScript to the site templates in order to trigger the / key Built its own little Python CORS web server to capture JSON data Rewrote the template to capture that data and send it to the server Scripted its way through the Web Component shadow DOM to the information it needed Opened Safari to confirm the source of the bug Modified its custom template to hack in a potential fix Confirmed the hacked fix worked Reported back on how to fix the problem Like I said, relentlessly proactive! An estimate of the cost I'm currently on the $100/month Claude Max plan, which includes a generous allowance for Fable up until June 22nd after which Anthropic say they'll start charging full API prices for it. I'm using AgentsView to track my spending (see this TIL ). Here's what AgentsView says this session would have cost me if I was paying full price for it: ~ % uvx agentsview session usage be8850a7-6119-46a0-b5d6-79c7fff5ae2b Session: be8850a7-6119-46a0-b5d6-79c7fff5ae2b Agent: claude Output: 68606 Peak ctx: 113178 Cost: ~$12.11 (claude-fable-5, claude-opus-4-8) If you don't keep a close eye on it, Fable will quite happily burn $12 in tokens inventing new ways to debug your CSS. I really need to lock this thing down On the one hand, watching Fable go to extreme lengths to get the information that it needed to debug what was, in the end, a two-line CSS fix, was fascinating . But on the other hand... this is a robust reminder that coding agents can do anything you can do by typing commands into a terminal - and frontier models know every trick in the book, and evidently a few that nobody has ever written down before. If Fable had been acting on malicious instructions - a prompt injection attack hidden in code or an issue thread, or something I'd carelessly pasted into my terminal - it's alarming to think quite how far it could go to exfiltrate data or cause other forms of mischief. Running coding agents outside of a sandbox has always been a bad idea - it's my top contender for a Challenger disaster incident, as described by Johann Rehberger in The Normalization of Deviance in AI . Fable is arguably smarter and hence more suspicious of potentially malicious instructions. But that smartness is very much a two-edged sword: if it does get subverted by instructions, the amount of damage it can do given its relentless proactivity is terrifying. Tags: ai , prompt-injection , generative-ai , llms , ai-assisted-programming , coding-agents , claude-code , claude-mythos
Release: datasette 1.0a33 This alpha is a significant step on the road to a stable 1.0, finally extending the ?_extra= pattern I introduced in Datasette 1.0a3 to cover queries and rows in addition to tables. That pattern is also now documented ! I wrote a whole lot more about the new release on the Datasette project blog: Datasette 1.0a33 with JSON extras in the API . Because API explorer tools are almost free to build now I had Claude Fable 5 in Claude Code (for the plan ) and GPT-5.5 xhigh in Codex Desktop (for the implementation ) build me this custom extras API explorer to help demonstrate the feature: Tags: projects , datasette , annotated-release-notes , ai-assisted-programming
Release: asyncinject 0.7 I built this utility library to support an asyncio dependency injection pattern a few years ago. I was using it with Datasette and Claude Fable 5 spotted some bugs in the dependency which it then fixed for me. It's a very proactive model! Tags: async , projects , python , claude-mythos
Anthropic Walks Back Policy That Could Have ‘Sabotaged’ AI Researchers Using Claude Big scoop for Maxwell Zeff at Wired: “We’re changing Fable 5’s safeguards for frontier LLM development to make them visible.” Anthropic said in a statement to WIRED. “We made the wrong tradeoff and we apologize for not getting the balance right.” There's been a huge outcry about Anthropic's policy, tucked away in their system card , that Claude Fable/Mythos would identify "requests targeting frontier LLM development" and "limit effectiveness" without notifying the user. It's good news that they're dropping the invisible aspect of this. It would be a whole lot better of they dropped this category of refusals entirely. Update : More details from @ClaudeDevs on Twitter : We’re rolling out changes to make Fable 5’s safeguards for frontier LLM development visible. Starting this week, flagged requests will visibly fall back to Opus 4.8—the same as our safeguards for cyber and bio. You will see this every time it happens. On the API, any flagged requests will return a reason for their refusal (coming to server-side fallback in the next few days). We wanted to deploy Fable 5 to our users quickly and safely. Visible safeguards can be probed, so they have to be robust, which takes time to get right. Invisible safeguards can be targeted more narrowly, allowing us to ship quickly with very few false positives. We went with invisible safeguards for this reason—and that was the wrong tradeoff. You should have visibility into the safeguards we have in place, and why. We’re sorry for not getting the balance right. Via @zeffmax Tags: ai , generative-ai , llms , anthropic , claude , ai-ethics , claude-mythos
OpenAI supports the EU Code of Practice on AI content transparency, advancing provenance standards and tools to help people understand AI-generated content.
Discover how astrophysicist Chi-kwan Chan uses Codex to build black hole simulations, helping scientists study extreme physics and test Einstein’s theory of general relativity.
Release: datasette-agent 0.2a0 Highlights from the release notes: Tools can now ask the user questions mid-execution. Tools that declare a context parameter receive a ToolContext object, and await context.ask_user(...) can ask a yes/no, multiple-choice ( options=[...] ) or free-text ( free_text=True ) question. While a question is unanswered the agent turn suspends: the question renders as a form in the chat UI and persists to the internal database, so suspended conversations survive a server restart. Once answered, the tool re-executes from the top with stored answers replayed, so call ask_user() before performing side effects. #20 New built-in save_query tool: the agent can save SQL it has written as a Datasette stored query . Saving always requires human approval - the agent shows the full SQL plus the proposed name, database and visibility, and nothing is stored until you click Yes. #20 The ask_user() feature was enabled by the new LLM alpha I built yesterday with the help of Claude Fable 5. Tags: ai , datasette , generative-ai , llms , datasette-agent
DiffusionGemma Last May Google briefly released an experimental Gemini Diffusion model. I tried the preview at the time and recorded it running at 857 tokens/second. It was an exciting model, but Google made no further announcements about it. That research has returned in the best possible way: as a new open weight (Apache 2 licensed) Gemma model, google/diffusiongemma-26B-A4B-it . NVIDIA are currently hosting the model for free on their NIM cloud API. I used that API to generate this pelican , which took 4.4s (according to time uv run generate.py ) to return 2,409 tokens - so at least 500 tokens/second. Via Hacker News Tags: google , ai , generative-ai , llms , nvidia , pelican-riding-a-bicycle , gemma , llm-release , llm-performance
Easy solution to slow down recursive AI self improvement: The lab with the top-ranked model must agree THEY must not use it for working on frontier AI But everyone else should have access to it. By definition, this means the frontier doesn't advance. It also has the critical benefit of avoiding a dangerous power imbalance. Anthropic has chosen the opposite of the safe path: they are allowing themselves, the current top lab, to use their top model for frontier AI research. They've said they'll sabotage others who try. This means the AI frontier advances, & power imbalance increases. (To be clear, I don't think we should try to slow down recursive AI self improvement - I think we should open it up and democratize it as much as possible. My point is: if you claim we should slow down, and you have the best model, you should ensure your org can't use it.) — Jeremy Howard , in a Twitter thread Tags: ai-ethics , anthropic , generative-ai , claude-mythos , jeremy-howard , ai , llms
Application Services for Private Origins is available now in closed beta. Route public hostnames to private IP origins over your existing IPsec, GRE, CNI, or Cloudflare Mesh paths. No public IPs or extra connector software required.
A new report from OpenAI details PRC-linked influence operations using AI to target U.S. tech debates, data center narratives, tariffs, and false claims about ChatGPT.