Why the ‘glorified host’ for AI is exactly the Kubernetes we need

Summary:

I recently caught a post from Hyperframe Research that asked a question many of us in the cloud-native trenches have

The post Why the ‘glorified host’ for AI is exactly the Kubernetes we need appeared first on The New Stack.

Major SteamOS update adds support for Steam Machine, even more third-party hardware

Summary:
Both AMD- and Intel-based hardware is getting better support in SteamOS 3.8.

A new anonymous Substack alleges AI compliance startup Delve "faked" compliance for startups by generating pre-populated audit reports and fabricating evidence (DeepDelver)

Summary:

DeepDelver:
A new anonymous Substack alleges AI compliance startup Delve “faked” compliance for startups by generating pre-populated audit reports and fabricating evidence  —  How Delve managed to falsely convince hundreds of customers they were compliant and then lied about it when exposed and called out

World’s most efficient EV truck charger delivers 1 MW power with 99.26% efficiency

Summary:
German researchers have recently unveiled a new megawatt EV charging system for heavy-duty transport that...

Google is running a "small" experiment replacing news headlines in search results with AI-generated ones, after adding the feature in Google Discover in January (Sean Hollister/The Verge)

Summary:

Sean Hollister / The Verge:
Google is running a “small” experiment replacing news headlines in search results with AI-generated ones, after adding the feature in Google Discover in January  —  We're seeing Verge headlines rewritten by Google AI. … Since roughly the turn of the millennium, Google Search has been the bedrock of the web.

OpenAI acquires Astral to bring open source Python developer tools to Codex — but details are still fuzzy

Summary:

OpenAI this week announced its acquisition of Astral to bring the startup’s open source Python developer tools into the Codex

The post OpenAI acquires Astral to bring open source Python developer tools to Codex — but details are still fuzzy appeared first on The New Stack.

Mistral's CEO proposes a revenue-based levy for AI model providers in the EU, to be invested "in new content creation and supporting Europe's cultural sectors" (Arthur Mensch/Financial Times)

Summary:

Arthur Mensch / Financial Times:
Mistral's CEO proposes a revenue-based levy for AI model providers in the EU, to be invested “in new content creation and supporting Europe's cultural sectors”  —  A revenue-based charge would protect the livelihoods of copyright holders and bring legal certainty

Why flat Kubernetes networks fail at scale

Summary:

Rethinking network security hierarchies for cloud-native platforms Kubernetes networking is powerful. Its flexibility lets teams connect hundreds of microservices across

The post Why flat Kubernetes networks fail at scale appeared first on The New Stack.

World’s first high-endurance drone vessel deploys anti-submarine warfare glider on its own

Summary:
An unmanned surface vessel has launched an ocean glider autonomously. ZeroUSV’s vessel deployed a Slocum...

The White House releases an AI policy framework, explicitly calling on Congress to preempt state AI laws, create age-gating requirements for AI models, and more (Politico)

Summary:

Politico:
The White House releases an AI policy framework, explicitly calling on Congress to preempt state AI laws, create age-gating requirements for AI models, and more  —  The framework explicitly calls on Congress to preempt any state laws that regulate the way models are developed or that penalize companies …

New 3D-printed robot uses wind instead of batteries to explore harsh terrains

Summary:
A new low-cost, 3D-printed robot has been designed to traverse harsh environments using nothing but...

From pillars to platform: How open observability data is changing the industry

Summary:

As the observability industry pivots from proprietary systems to open frameworks, the traditional “pillars” of metrics, logs, and traces are

The post From pillars to platform: How open observability data is changing the industry appeared first on The New Stack.

Humanoid robot’s dance gone wrong causes chaos at busy California restaurant

Summary:
A service robot meant to entertain diners caused a sudden disruption at a restaurant in...

Filing: Unitree, China's leading robot company, is aiming to raise ~$610M in its Shanghai IPO; it brought in ~$247M in revenue in 2025 and ~$41.6M in net profit (Jessica Sui/Bloomberg)

Summary:

Jessica Sui / Bloomberg:
Filing: Unitree, China's leading robot company, is aiming to raise ~$610M in its Shanghai IPO; it brought in ~$247M in revenue in 2025 and ~$41.6M in net profit  —  China's leading robot company, Unitree Robotics, has filed for an initial public offering in Shanghai, moving a step closer …

Dark Eagle: US Army set to deploy its first hypersonic missile with 2,175-mile-range

Summary:
Overcoming prolonged delays and technical challenges, the US Army’s Long-Range Hypersonic Weapon is approaching its...

Brendan Carr Crafting ‘Patriotic’ Call Center Onshoring Plan To Provide Cover For Mass Looming Telecom Layoffs

Summary:
When he’s not busy trampling free speech, crushing the First Amendment, and destroying media consolidation and consumer protection standards, Brendan Carr has other hobbies. Like helping the telecom industry patriotically sell a brutal coming wave of new layoffs caused by the kind of industry consolidation he regularly rubber stamps. Carr recently began circulating plans for […]

54,300-ton Viking Libra: World’s first hydrogen-powered cruise ship floats out

Summary:
The world’s first hydrogen-powered ship was floated out recently at the Ancona Shipyard in Italy,...

Humanoid robot masters tennis with 96.5% accuracy using simplified human motion

Summary:
Researchers in China have developed a new system that significantly improves how humanoid robots learn...

Linux kernel scale is swamping an already-flawed CVE system

Summary:

Linux kernel developers operate under constraints that very few other open source community maintainers experience. Evolving the capabilities of the

The post Linux kernel scale is swamping an already-flawed CVE system appeared first on The New Stack.

Reinforced solar cells with 26% efficiency can work even in tough space conditions

Summary:
In space, satellites face brutal temperature shifts from hot sunlight to freezing shadows. For the...

Sources: Amazon's ZeroOne unit is developing "Transformer", a phone that syncs with Alexa; it would be Amazon's second smartphone, after the Fire Phone in 2014 (Greg Bensinger/Reuters)

Summary:

Greg Bensinger / Reuters:
Sources: Amazon's ZeroOne unit is developing “Transformer”, a phone that syncs with Alexa; it would be Amazon's second smartphone, after the Fire Phone in 2014  —  In 2014, Amazon (AMZN.O) introduced its first smartphone, hoping to take on Apple and Samsung.

Shareholder letter: crypto exchange Gemini has cut ~30% of its staff since the start of 2026 and employed ~445 people as of March 1; Gemini lost $500M+ in 2025 (Olga Kharif/Bloomberg)

Summary:

Olga Kharif / Bloomberg:
Shareholder letter: crypto exchange Gemini has cut ~30% of its staff since the start of 2026 and employed ~445 people as of March 1; Gemini lost $500M+ in 2025  —  Gemini Space Station Inc.'s workforce reduction has reached roughly 30% since the start of the year following layoffs …

Memo: ByteDance agrees to sell gaming unit Moonton to Savvy Games, owned by Saudi Arabia's PIF, sources say for $6B; ByteDance bought Moonton for ~$4B in 2021 (Bloomberg)

Summary:

Bloomberg:
Memo: ByteDance agrees to sell gaming unit Moonton to Savvy Games, owned by Saudi Arabia's PIF, sources say for $6B; ByteDance bought Moonton for ~$4B in 2021  —  Savvy Games Group has agreed to buy the Moonton studio from ByteDance Ltd. as the Saudi Arabian firm builds up its mobile gaming assets.

Xiaomi unveils its MiMo-V2 AI models, including the 1T-parameter MiMo-V2-Pro, codenamed Hunter Alpha, which Xiaomi says benchmarks close to GPT-5.2 and Opus 4.6 (Carl Franzen/VentureBeat)

Summary:

Carl Franzen / VentureBeat:
Xiaomi unveils its MiMo-V2 AI models, including the 1T-parameter MiMo-V2-Pro, codenamed Hunter Alpha, which Xiaomi says benchmarks close to GPT-5.2 and Opus 4.6  —  Led by Fuli Luo, a veteran of the disruptive DeepSeek R1 project, the release represents what Luo characterizes as a “quiet ambush” on the global frontier.

Israel strikes Iran’s warships, naval base in Caspian Sea for first time in war

Summary:
Israel has carried out a rare long-range strike on Iranian naval assets in the Caspian...

It's real, enterprises just need to do the CISO work and SRE work

Summary:

After using Claude more and more for task in my personal life, my current zinger analyst take on the Squawk Box would be: “OpenAI talks about business strategy, Anthropic just does it.” It’s really getting close to a sci-fi personal assistant. It takes A LOT of work to get your rig (or “harness”) setup, and to continually tune it, but it’s amazing.

Once CEO’s get their hands on this for a week, and the IT departments and CISOs figure it out, it’ll be amazing in enterprise life…again, not just for programmers, but for everyone.

The biggest benefit will be less meetings, less having to be a meat-mouse for executives who can’t or won’t make their own slides, and finally getting CRM and ERP “business analytics” nailed.

You know, until we get used to productivity benefits in 6 months and then just hate it like we do all enterprise software, because we want the next impossible thing.

The SaaS Observability Era is Ending: Why BYOC Is the Future of Telemetry

Summary:

Filing: the SEC approved a Nasdaq rule change allowing some securities to trade in tokenized form; the new approach will be tested in an upcoming pilot program (The Block)

Summary:

The Block:
Filing: the SEC approved a Nasdaq rule change allowing some securities to trade in tokenized form; the new approach will be tested in an upcoming pilot program  —  Quick Take  — Eligible participants will be able to settle trades in tokenized form during a pilot program operated by the Depository Trust Company.

Alibaba and Tencent lost a combined $66B in market value in ~24 hours after investors punished the two Chinese tech leaders for unclear AI monetization plans (Bloomberg)

Summary:

Bloomberg:
Alibaba and Tencent lost a combined $66B in market value in ~24 hours after investors punished the two Chinese tech leaders for unclear AI monetization plans  —  Alibaba Group Holding Ltd. and Tencent Holdings Ltd. lost $66 billion of market value in roughly 24 hours, after the market punished …

Sources: Indian AI data center operator Yotta is aiming to secure ~$500M to $600M at a ~$4B valuation, then file initial paperwork for an IPO within weeks (Saritha Rai/Bloomberg)

Summary:

Saritha Rai / Bloomberg:

Microsoft releases MAI-Image-2, ranked third on Arena AI's text-to-image leaderboard behind only models from Google and OpenAI, available in the MAI Playground (Jose Antonio Lanz/Decrypt)

Summary:

Jose Antonio Lanz / Decrypt:
Microsoft releases MAI-Image-2, ranked third on Arena AI's text-to-image leaderboard behind only models from Google and OpenAI, available in the MAI Playground  —  Microsoft has been quietly building its own image generator.  Announced Thursday by the company's AI Superintelligence team …

Filing: Blue Origin seeks US FCC approval to deploy nearly 52K solar-powered satellites as part of Project Sunrise, its proposed orbital AI data center system (Micah Maidenberg/Wall Street Journal)

Summary:

Micah Maidenberg / Wall Street Journal:
Filing: Blue Origin seeks US FCC approval to deploy nearly 52K solar-powered satellites as part of Project Sunrise, its proposed orbital AI data center system  —  Blue Origin joins SpaceX and Starcloud in filing FCC applications for AI satellites&nb

Greater Than Zero: The Anti-AI Pushback On Gaming Preservation Efforts Makes No Sense

Summary:
There is an old axiom you will have heard of before: don’t let the perfect be the enemy of the good. If we wanted to boil this down to a math equation, it might be described as something like: 0 < any positive integer. It’s not a difficult concept to grasp, typically, until you add […]

Sources: Jeff Bezos' $100B fund would be part of the same holding company as Project Prometheus and would invest in companies that could benefit from its tech (New York Times)

Summary:

New York Times:
Sources: Jeff Bezos' $100B fund would be part of the same holding company as Project Prometheus and would invest in companies that could benefit from its tech  —  The new fund would operate alongside the Amazon founder's A.I. start-up, Project Prometheus.  —  Jeff Bezos is in early discussions …

templUI brings shadcn-style components to Go and templ

Summary:

#​594 — March 20, 2026 Read the Web Version Go Weekly

Ctrl-Alt-Speech: Money For Nothing And Clicks For A Fee

Summary:
Ctrl-Alt-Speech is a weekly podcast about the latest news in online speech, from Mike Masnick and Everything in Moderation‘s Ben Whitelaw. Subscribe now on Apple Podcasts, Overcast, Spotify, Pocket Casts, YouTube, or your podcast app of choice — or go straight to the RSS feed. In this week’s round-up of the latest news in online […]

Migrating Etsy’s database sharding to Vitess

Summary:

Etsy has maintained a sharded MySQL architecture since around 2010. This database cluster contains most of Etsy’s online data and is made up of ~1,000 tables distributed across ~1,000 shards. Over the last 16 years, it has grown significantly: combined, these tables have over 425 TB of data and receive roughly 1.7 million requests per second.

Etsy engineers access our MySQL data through a proprietary object-relational mapping (ORM). The ORM has a corresponding model for each MySQL table. When a table is sharded, its rows are partitioned among different databases known as shards. Each shard has identical table schemas and contains a distinct subset of all rows. Sharded tables require a unique id field to be designated on the model as the “shardifier id” which determines the shard for each record.

Shardifier ids are designed to co-locate related data in order to minimize the number of databases we need to rely on for a given query. For example, all records related to a single shop or user are

The Government Uses Targeted Advertising to Track Your Location. Here’s What We Need to Do.

Summary:
We’ve all had the unsettling experience of seeing an ad online that reveals just how much advertisers know about our lives. You’re right to be disturbed. Those very same online ad systems have been used by the government to warrantlessly track peoples’ locations, new reporting has confirmed. For years, the internet advertising industry has been […]

Secure Code Warrior AI Agent Applies Policies to AI Generated Code

Summary:

Powering the agents: Workers AI now runs large models, starting with Kimi K2.5

Summary:
Kimi K2.5 is now on Workers AI, helping you power agents entirely on Cloudflare’s Developer Platform. Learn how we optimized our inference stack and reduced inference costs for internal agent use cases.

Prediction Markets Promised Better Information. Instead They’re Creating Powerful Incentives to Corrupt Information.

Summary:
There’s a concept in economics known as Goodhart’s Law, often summarized as: “When a measure becomes a target, it ceases to be a good measure.” The idea, originally about monetary policy, has proven remarkably durable across domains. When you attach high enough stakes to a single metric, people stop trying to accurately reflect reality and […]

It was a rather stressful period

Summary:

Related to your interests

• Your database is about to become an AI tool. Is it ready?

• Banks struggle to scale AI as legacy tech devours IT budgets - “Executives said only 29% of annual IT budgets are set aside for transformative technologies while a larger chunk - 43% - is devoted to maintaining legacy systems.”

• The AI-Washing of Job Cuts Is Corrosive and Confusing

• Hardened containers for AI - “Lorenc was blunt: The bottleneck in modern software isn’t generating code anymore; it’s trust. In his keynote, he describes how AI is collapsing exploit development tim

How Slack Rebuilt Notifications 📣

Summary:
Introduction 🔔  At Slack, notifications are how teams stay in the loop, but they can also become overwhelming when not designed with intention. Our goal was to make staying informed feel effortless. We set out to rebuild one of Slack’s most complicated systems from the ground up by bringing calm, consistency, and clarity to the…

Backend Development in 2026: What’s Changed, What Matters, and What to Learn Next

Summary:

Overview of Backend Development Backend development, often called server-side development, involves building and overseeing applications that run on computers controlled

The post Backend Development in 2026: What’s Changed, What Matters, and What to Learn Next appeared first on The New Stack.

Jellyfish AI development study: The real sting has yet to land

Summary:

Developers are adopting and eschewing AI coding tools depending on their trust in the new automation mantra, the level of

The post Jellyfish AI development study: The real sting has yet to land appeared first on The New Stack.

OpenAI Bolsters AI Coding with Acquisition of Python Toolmaker Astral

Summary:

White South Africans would Rather Live In South Africa Than In The US Under Trump

Summary:
Shortly after Trump took office for a second time, his administration made it clear that it felt constitutional rights were merely privileges it would extend only to those who fully supported whatever the hell the administration happened to be doing. At the time, the administration was not only engaged in a full-blown, bigoted war against […]

Meta decides not to kill Horizon Worlds VR after all

Summary:
VR will be on life support while mobile remains the focus, though.

Daily Deal: Zeus Smart Car Kit for Arduino (Battery & SunFounder UNO R3 Included)

Summary:
Discover the innovative Zeus Smart Car Kit, a versatile robot equipped with 4WD omnidirectional movement, FPV (First Person View), app-based remote control, RGB lighting, and a durable metal frame. Powered by an Arduino Uno-compatible main board, this all-in-one kit delivers multi-functional capabilities, including obstacle avoidance, line tracking, IR remote control, face detection, gesture recognition, voice […]

Google details new 24-hour process to sideload unverified Android apps

Summary:
The "advanced flow" will be available before verification enforcement begins later this year.

Despite hardware limits, Parallels supports running Windows on MacBook Neo

Summary:
"8GB unified memory is the minimum practical configuration."

Lineaje Adds Ability to Automatically Apply Governance Policies to AI Components

Summary:

Afroman Wins: Jury Rules Mocking Cops Who Raided Your Home Is Protected Speech

Summary:
As we wrote just yesterday, the defamation trial brought by seven Adams County, Ohio deputies against rapper Afroman was going about as well for the officers as their original botched raid on his home. Today we can report the inevitable conclusion: the jury sided entirely with Afroman, clearing him of all liability after just hours […]

Cursor’s Composer 2 beats Opus 4.6 on coding benchmarks at a fraction of the price

Summary:

Cursor on Thursday released Composer 2, the third generation of its in-house coding model. The model outperforms Anthropic’s Opus 4.6

The post Cursor’s Composer 2 beats Opus 4.6 on coding benchmarks at a fraction of the price appeared first on The New Stack.

Sampling: the philosopher’s stone of distributed tracing

Summary:

In modern observability, distributed tracing is often considered the most expressive signal. It can be used to capture much of

The post Sampling: the philosopher’s stone of distributed tracing appeared first on The New Stack.

Checkmarx Adds Orchestration Framework to DevSecOps Platform

Summary:

SpyCloud’s 2026 Identity Exposure Report Reveals Explosion of Non-Human Identity Theft

Summary:

David Zaslav Will Soon Get $550 Million For Disastrous Warner Bros Reign

Summary:
Brunchlords can only fail upward. It’s physics. Warner Brothers CEO David Zaslav is poised to get as much as $550 million in compensation and tax reimbursement as the company prepares to be acquired by Larry Ellison’s CBS/Paramount: “Zaslav, president and CEO of Warner Bros. Discovery, is set to receive $34.2 million in cash severance; $517.2 […]

Building a Kubernetes-native pattern for AI infrastructure at scale

Summary:

When AI developer and ops teams first adopt large models on Kubernetes, the focus is usually on getting something to

The post Building a Kubernetes-native pattern for AI infrastructure at scale appeared first on The New Stack.

TCL’s German QLED ban puts pressure on TV brands to be more honest about QDs

Summary:
"This is a serious warning shot."

Open SWE Captures the Architecture That Stripe, Coinbase and Ramp Built Independently for Internal Coding Agents

Summary:

U.S. District Court Issues Preliminary Injunction Against RFK, HHS For Its Vaccine Schedule Changes

Summary:
It was mere days ago that we were discussing an interesting lawsuit brought by the American Academy of Pediatrics, among others, challenging RFK Jr. and HHS for violating the Administrative Procedures Act in making changes to the CDC’s ACIP panel and immunization schedules. If you’re not up on what the APA is and does, the […]

Summary:

Internet Coin-op

Arcjet Extends Runtime Policy Engine to Block Malicious Prompts

Summary:

Google, Microsoft and Peers Donate to Support Overloaded Open Source Maintainers

Summary:

Hacking Workshop for April/May 2026

Summary:

I'm planning to hold a single hacking workshop for April and May combined, covering Masahiko Sawada's talk, Breaking away from FREEZE and Wraparound, given at PGCon 2022. If you're interested in joining us, please sign up using this form and I will send you an invite to one of the sessions. Thanks to Sawada-san for agreeing to join us.

AI-powered event response for Amazon EKS

Summary:
In this post, you'll learn how AWS DevOps Agent integrates with your existing observability stack to provide intelligent, automated responses to system events.

Friend Bubbles: Enhancing Social Discovery on Facebook Reels

Summary:

Friend bubbles in Facebook Reels highlight Reels your friends have liked or reacted to, helping you discover new content and making it easier to connect over shared interests. This article explains the technical architecture behind friend bubbles, including how machine learning estimates relationship strength and ranks content your friends have interacted with to create more [...]

Read More...

The post Friend Bubbles: Enhancing Social Discovery on Facebook Reels appeared first on Engineering at Meta.

Spacelift Intelligence Vibe-Codes Infrastructure

Summary:

The AI revolution will be open-sourced

Summary:

At CES 2026, NVIDIA CEO Jensen Huang made the case that AI will proliferate when open innovation is activated across

The post The AI revolution will be open-sourced appeared first on The New Stack.

Java 26 lands without an LTS badge. Here’s why developers should care anyway.

Summary:

released Java 26 on Tuesday, the opening day of its JavaOne 2026 conference in Redwood Shores, Calif., continuing the platform’s

The post Java 26 lands without an LTS badge. Here’s why developers should care anyway. appeared first on The New Stack.

Chainguard has a fix for the open source packages your AI agents keep grabbing

Summary:

As AI coding assistants and autonomous agents dramatically accelerate software development, they’re introducing a security problem that most organizations aren’t

The post Chainguard has a fix for the open source packages your AI agents keep grabbing appeared first on The New Stack.

Sauce Labs wants to solve an AI-created problem nobody wanted to work on

Summary:

AI code-generation tools have compressed software development cycles, but testing hasn’t kept up. Sauce Labs thinks it has found an

The post Sauce Labs wants to solve an AI-created problem nobody wanted to work on appeared first on The New Stack.

Capital One deprecated an AI tool it once championed. Its DevEx chief says that’s the point.

Summary:

How could you possibly ensure you are providing the right tools and processes to 14,000 engineers? How do you even

The post Capital One deprecated an AI tool it once championed. Its DevEx chief says that’s the point. appeared first on The New Stack.

Introducing Custom Regions for precision data control

Summary:
We are expanding Regional Services with new pre-defined regions and the launch of Custom Regions. Customers can now define precise geographical boundaries for data processing, tailored to meet their compliance and performance needs.

Scaling Btrfs to petabytes in production: a 74% cost reduction story

Summary:

At Chronosphere, we saved 74% of our storage costs by moving petabytes of time-series data from ext4 to Btrfs, the

The post Scaling Btrfs to petabytes in production: a 74% cost reduction story appeared first on The New Stack.

Why your observability bill keeps growing (and it’s not your vendor’s fault)

Summary:

Over the past year, I have spoken with dozens of engineering teams about their observability pipelines and analyzed their production

The post Why your observability bill keeps growing (and it’s not your vendor’s fault) appeared first on The New Stack.

Ranking Engineer Agent (REA): The Autonomous AI Agent Accelerating Meta’s Ads Ranking Innovation

Summary:

Meta’s Ranking Engineer Agent (REA) autonomously executes key steps across the end-to-end machine learning (ML) lifecycle for ads ranking models. This post covers REA’s ML experimentation capabilities: autonomously generating hypotheses, launching training jobs, debugging failures, and iterating on results. Future posts will cover additional REA capabilities. REA reduces the need for manual intervention. It manages [...]

Read More...

The post Ranking Engineer Agent (REA): The Autonomous AI Agent Accelerating Meta’s Ads Ranking Innovation appeared first on Engineerin

Chainguard thinks most DevOps teams are solving container security the hard way

Summary:

Chainguard OS is great, but what if you want to customize your own Linux? Now, with the Chainguard OS Package,

The post Chainguard thinks most DevOps teams are solving container security the hard way appeared first on The New Stack.

OpenAI’s GPT-5.4 mini and nano are built for the subagent era

Summary:

On Tuesday, OpenAI released GPT-5.4 mini and nano, two smaller models designed for tasks that agentic AI systems delegate: codebase searches,

The post OpenAI’s GPT-5.4 mini and nano are built for the subagent era appeared first on The New Stack.

WebMCP turns any Chrome web page into an MCP server for AI agents

Summary:

About two weeks ago, I posted about websites returning Markdown to make life easier for AI agents. While agents are already quite

The post WebMCP turns any Chrome web page into an MCP server for AI agents appeared first on The New Stack.

GitHub Copilot’s effect on collaboration has stunned researchers

Summary:

Generative AI tools like GitHub Copilot are not just speeding up coding; they are quietly rewiring how software developers divide

The post GitHub Copilot’s effect on collaboration has stunned researchers appeared first on The New Stack.

How we optimized Dash's relevance judge with DSPy

Summary:
We used DSPy to turn prompt engineering for our relevance judge into a measurable, automated optimization loop, improving task performance, cost, and how reliably it works in production.

The security hole that every enterprise AI deployment has (but nobody looks for)

Summary:

The anti-automation glitterati may now be preparing for a new swathe of naysaying as agentic AI services are brought to

The post The security hole that every enterprise AI deployment has (but nobody looks for) appeared first on The New Stack.

After three months, Samsung is ending sales of the $2,899 Galaxy Z TriFold

Summary:
Samsung didn't offer an explanation for its decision, but it's not exactly a surprise.

What is KubeVirt and why it’s growing

Summary:

KubeVirt is an open-source project that brings virtual machines into the Kubernetes control plane, letting teams run VMs and containers

The post What is KubeVirt and why it’s growing appeared first on The New Stack.

Switch 2's new "Handheld Mode Boost" can run original Switch games at 1080p

Summary:
Switch games running at 720p can look worse on the Switch 2's 1080p display.

Samsung Galaxy S26 Ultra review: Private and performant

Summary:
If you want to spend $1,300 on a phone, it might as well be this one.

The AI blind spot debt: the hidden cost killing your innovation strategy

Summary:

In today’s AI rush, I’ve seen even the most disciplined organizations find it nearly impossible to enforce the hard-won lessons

The post The AI blind spot debt: the hidden cost killing your innovation strategy appeared first on The New Stack.

Firefighting drones head to Aspen—can they suppress a blaze before humans arrive?

Summary:
Seneca's drones carry foam, fly at night, and don't need an on-site pilot.

Managed OpenClaw bids to kill hidden token tax on AI agents

Summary:

Featherless is a serverless platform specialist that provides API-based access to open-source AI models via a supporting infrastructure. The concept

The post Managed OpenClaw bids to kill hidden token tax on AI agents appeared first on The New Stack.

Why agentic AI stalls in production — and how a control plane fixes it

Summary:

For SREs, platform engineers, and AI/ML teams, agentic AI makes many promises that are hard to ignore. Autonomous agents that

The post Why agentic AI stalls in production — and how a control plane fixes it appeared first on The New Stack.

AI Brain Fry, Zombie Projects, and Gentleman Vibe Coders - Related to your interests, Monday

Summary:

KubeCon EU is next week, in Amsterdam, and I’m talking at VMUG Amsterdam tomorrow about what private equity does to enterprise software companies. If you’re at RAI, come say hi. AI might have written some of the below; we’re still working on keeping it under control for “helpful” content generation.

Related to your interests

• When Using AI Leads to Brain Fry - In my experience, it’s not so much AI doing this as AI being a tool that makes people so productive that they experience getting a lot of shit done. Being hyper-productive is exhausting, but in a good way like (so I am told) exercising is exhausting but ultimately “good.” // Related: ‘AI brain fry’ affects employees managing too many agents - Those who reported having to maintain a high degree of AI oversight reported spending 14 percent more mental ene

Making AI Write Android Code Our Way: A Practical Guide to Agent Skills

Summary:

Turning knowledge into reusable AI agent instructions for a small, fast-moving team.

We're a small Android team at Medium, just a handful of engineers maintaining and evolving the Medium Android app. Our codebase follows Clean Architecture with Kotlin, Jetpack Compose, Hilt, Apollo GraphQL, and a growing number of feature modules. Like most Android teams, we have strong opinions about how code should be structured: where ViewModels get their data, how analytics events flow, how feature flags are checked, what a "new screen" looks like from Fragment to preview function.

The problem? Those opinions lived in PR review comments, Slack threads, and the heads of engineers who'd been around long enough to know the patterns. When AI coding assistants arrived, they could generate Kotlin code but not our Kotlin code. The output was gener

Standing up for the open Internet: why we appealed Italy’s "Piracy Shield" fine

Summary:
Cloudflare is appealing a €14 million fine from Italian regulators over "Piracy Shield," a system that forces providers to block content without oversight. We are challenging this framework to protect the Internet from disproportionate overblocking and lack of due process.

Apple’s AirPods Max 2 bring H2 chip, boosted ANC in April for $549

Summary:
Apple's over-head headphones get an update after over five years.

COMMENT to the MCP Rescue

Summary:

The COMMENT command has been in Postgres for decades. It allows text descriptions to be attached to almost any database object. During its long history, it was mostly seen as a nice-to-have addition to database schemas, allowing administrators and developers to more easily understand the schema. Tools like pgAdmin allow you to assign and view comments on database objects.

Now, in the AI era, there is something else that needs to understand database schemas — MCP clients. Without database object comments, MCP clients can only use the database schemas, object names, and constraints. With database comment

Yahoo CEO Jim Lanzone on reviving the web’s homepage

Summary:
Today, I’m talking with Jim Lanzone, who is the CEO of Yahoo. It’s basically impossible to sum up the Yahoo story, but the short version of it is that a long time ago Yahoo paid Google to run the search box on its website, and basically everything has gone sideways since. You’ll hear Jim refer […]

Using Tools: A Meeting Scheduler

Summary:

Introduction

LLMs are great, but they are trained on public data sets. In some cases, you need the LLM to use data that’s not publicly available or that’s frequently changing. There are several ways to make such data available to LLMs:

• Tool/function calls
• Retrieval-augmented generation (aka RAG)
• MCP

In coding agents, you can also add skills.

In this post we’ll focus on function calling.

How Does It Work?

When interacting with an LLM, you can provide a description of available tools if the model supports tool calling. If the LLM reasons that the best answer is to use one of the tools, it will return a reply that contains a tool call with the parameters to use. Then you make the function call and return the answer back to the LLM.

Aether OS is a computer in a browser built for the AT Protocol

Summary:
Aether OS puts a full-fledged desktop in your browser that ties directly into the AT Protocol. That means it connects to your Bluesky account and other public records. It offers a pretty full suite of apps, 42 in total, covering text editing, task management, and social media. There's even a rudimentary tracker for making chiptunes, […]