Connect with us
DAPA Banner

Tech

Anthropic’s Claude Code Security is available now after finding 500+ vulnerabilities: how security leaders should respond

Published

on

Anthropic pointed its most advanced AI model, Claude Opus 4.6, at production open-source codebases and found a plethora of security holes: more than 500 high-severity vulnerabilities that had survived decades of expert review and millions of hours of fuzzing, with each candidate vetted through internal and external security review before disclosure.

Fifteen days later, the company productized the capability and launched Claude Code Security.

Security directors responsible for seven-figure vulnerability management stacks should expect a common question from their boards in the next review cycle. VentureBeat anticipates the emails and conversations will start with, “How do we add reasoning-based scanning before attackers get there first?”, because as Anthropic’s review found, simply pointing an AI model at exposed code can be enough to identify — and in the case of malicious actors, exploit — security lapses in production code.

The answer matters more than the number, and it is primarily structural: how your tooling and processes allocate work between pattern-based scanners and reasoning-based analysis. CodeQL and the tools built on it match code against known patterns.

Advertisement

Claude Code Security, which Anthropic launched February 20 as a limited research preview, reasons about code the way a human security researcher would. It follows how data moves through an application and catches flaws in business logic and access control that no rule set covers.

The board conversation security leaders need to have this week

Five hundred newly discovered zero-days is less a scare statistic than a standing budget justification for rethinking how you fund code security.

The reasoning capability Claude Code Security represents, and its inevitable competitors, need to drive the procurement conversation. Static application security testing (SAST) catches known vulnerability classes. Reasoning-based scanners find what pattern-matching was never designed to detect. Both have a role.

Anthropic published the zero-day research on February 5. Fifteen days later, they shipped the product. While it’s the same model and capabilities, it is now available to Enterprise and Team customers.

Advertisement

What Claude does that CodeQL couldn’t

GitHub has offered CodeQL-based scanning through Advanced Security for years, and added Copilot Autofix in August 2024 to generate LLM-suggested fixes for alerts. Security teams rely on it. But the detection boundary is the CodeQL rule set, and everything outside that boundary stays invisible.

Claude Code Security extends that boundary by generating and testing its own hypotheses about how data and control flow through an application, including cases where no existing rule set describes. CodeQL solves the problem it was built to solve: data-flow analysis within predefined queries. It tells you whether tainted input reaches a dangerous function.

CodeQL is not designed to autonomously read a project’s commit history, infer an incomplete patch, trace that logic into another file, and then assemble a working proof-of-concept exploit end to end. Claude did exactly that on GhostScript, OpenSC, and CGIF, each time using a different reasoning strategy.

“The real shift is from pattern-matching to hypothesis generation,” said Merritt Baer, CSO at Enkrypt AI, advisor to Andesite and AppOmni, and former Deputy CISO at AWS, in an exclusive interview with VentureBeat. “That’s a step-function increase in discovery power, and it demands equally strong human and technical controls.”

Advertisement

Three proof points from Anthropic’s published methodology show where pattern-matching ends and hypothesis generation begins.

Commit history analysis across files. GhostScript is a widely deployed utility for processing PostScript and PDF files. Fuzzing turned up nothing, and neither did manual analysis. Then Claude pulled the Git commit history, found a patch that added stack bounds checking for font handling in gstype1.c, and reversed the logic: if the fix was needed there, every other call to that function without the fix was still vulnerable. In gdevpsfx.c, a completely different file, the call to the same function lacked the bounds checking patched elsewhere. Claude built a working proof-of-concept crash. No CodeQL rule describes that bug today. The maintainers have since patched it.

Reasoning about preconditions that fuzzers can’t reach. OpenSC processes smart card data. Standard approaches failed here, too, so Claude searched the repository for function calls that are frequently vulnerable and found a location where multiple strcat operations ran in succession without length checking on the output buffer. Fuzzers rarely reached that code path because too many preconditions stood in the way. Claude reasoned about which code fragments looked interesting, constructed a buffer overflow, and proved the vulnerability.

Algorithm-level edge cases that no coverage metric catches. CGIF is a library for processing GIF files. This vulnerability required understanding how LZW compression builds a dictionary of tokens. CGIF assumed compressed output would always be smaller than uncompressed input, which is almost always true. Claude recognized that if the LZW dictionary filled up and triggered resets, the compressed output could exceed the uncompressed size, overflowing the buffer. Even 100% branch coverage wouldn’t catch this. The flaw demands a particular sequence of operations that exercises an edge case in the compression algorithm itself. Random input generation almost never produces it. Claude did.

Advertisement

Baer sees something broader in that progression. “The challenge with reasoning isn’t accuracy, it’s agency,” she told VentureBeat. “Once a system can form hypotheses and pursue them, you’ve shifted from a lookup tool to something that can explore your environment in ways that are harder to predict and constrain.”

How Anthropic validated 500+ findings

Anthropic placed Claude inside a sandboxed virtual machine with standard utilities and vulnerability analysis tools. The red team didn’t provide any specialized instructions, custom harnesses, or task-specific prompting. Just the model and the code.

The red team focused on memory corruption vulnerabilities because they’re the easiest to confirm objectively. Crash monitoring and address sanitizers don’t leave room for debate. Claude filtered its own output, deduplicating and reprioritizing before human researchers touched anything. When the confirmed count kept climbing, Anthropic brought in external security professionals to validate findings and write patches.

Every target was an open-source project underpinning enterprise systems and critical infrastructure. Small teams maintain many of them, staffed by volunteers, not security professionals. When a vulnerability sits in one of these projects for a decade, every product that pulls from it inherits the risk.

Advertisement

Anthropic didn’t start with the product launch. The defensive research spans more than a year. The company entered Claude in competitive Capture-the-Flag events where it ranked in the top 3% of PicoCTF globally, solved 19 of 20 challenges in the HackTheBox AI vs Human CTF, and placed 6th out of 9 teams defending live networks against human red team attacks at Western Regional CCDC.

Anthropic also partnered with Pacific Northwest National Laboratory to test Claude against a simulated water treatment plant. PNNL’s researchers estimated that the model completed adversary emulation in three hours. The traditional process takes multiple weeks.

The dual-use question security leaders can’t avoid

The same reasoning that finds a vulnerability can help an attacker exploit one. Frontier Red Team leader Logan Graham acknowledged this directly to Fortune’s Sharon Goldman. He told Fortune the models can now explore codebases autonomously and follow investigative leads faster than a junior security researcher.

Gabby Curtis, Anthropic’s communications lead, told VentureBeat in an exclusive interview the company built Claude Code Security to make defensive capabilities more widely available, “tipping the scales towards defenders.” She was equally direct about the tension: “The same reasoning that helps Claude find and fix a vulnerability could help an attacker exploit it, so we’re being deliberate about how we release this.”

Advertisement

In interviews with more than 40 CISOs across industries, VentureBeat found that formal governance frameworks for reasoning-based scanning tools are the exception, not the norm. The most common responses are that the area was considered so nascent that many CISOs didn’t think this capability would arrive so early in 2026.

The question every security director has to answer before deploying this: if I give my team a tool that finds zero-days through reasoning, have I unintentionally expanded my internal threat surface?

“You didn’t weaponize your internal surface, you revealed it,” Baer told VentureBeat. “These tools can be helpful, but they also may surface latent risk faster and more scalably. The same tool that finds zero-days for defense can expose gaps in your threat model. Keep in mind that most intrusions don’t come from zero-days, they come from misconfigurations.”

“In addition to the access and attack path risk, there is IP risk,” she said. “Not just exfiltration, but transformation. Reasoning models can internalize and re-express proprietary insights in ways that blur the line between use and leakage.”

Advertisement

The release is deliberately constrained. Enterprise and Team customers only, through a limited research preview. Open-source maintainers apply for free expedited access. Findings go through multi-stage self-verification before reaching an analyst, with severity ratings and confidence scores attached. Every patch requires human approval.

Anthropic also built detection into the model itself. In a blog post detailing the safeguards, the company described deploying probes that measure activations within the model as it generates responses, with new cyber-specific probes designed to track potential misuse. On the enforcement side, Anthropic is expanding its response capabilities to include real-time intervention, including blocking traffic it detects as malicious.

Graham was direct with Axios: the models are extremely good at finding vulnerabilities, and he expects them to get much better still. VentureBeat asked Anthropic for the false-positive rate before and after self-verification, the number of disclosed vulnerabilities with patches landed versus still in triage, and the specific safeguards that distinguish attacker use from defender use. The lead researcher on the 500-vulnerability project was unavailable, and the company declined to share specific attacker-detection mechanisms to avoid tipping off threat actors.

“Offense and defense are converging in capability,” Baer said. “The differentiator is oversight. If you can’t audit and bound how the tool is used, you’ve created another risk.”

Advertisement

That speed advantage doesn’t favor defenders by default. It favors whoever adopts it first. Security directors who move early set the terms.

Anthropic isn’t alone. The pattern is repeating.

Security researcher Sean Heelan used OpenAI’s o3 model with no custom tooling and no agentic framework to discover CVE-2025-37899, a previously unknown use-after-free vulnerability in the Linux kernel’s SMB implementation. The model analyzed over 12,000 lines of code and identified a race condition that traditional static analysis tools consistently missed because detecting it requires understanding concurrent thread interactions across connections.

Separately, AI security startup AISLE discovered all 12 zero-day vulnerabilities announced in OpenSSL’s January 2026 security patch, including a rare high-severity finding (CVE-2025-15467, a stack buffer overflow in CMS message parsing that is potentially remotely exploitable without valid key material). AISLE co-founder and chief scientist Stanislav Fort reported that his team’s AI system accounted for 13 of the 14 total OpenSSL CVEs assigned in 2025. OpenSSL is among the most scrutinized cryptographic libraries on the planet. Fuzzers have run against it for years. The AI found what they were not designed to find.

The window is already open

Those 500 vulnerabilities live in open-source projects that enterprise applications depend on. Anthropic is disclosing and patching, but the window between discovery and adoption of those patches is where attackers operate today.

Advertisement

The same model improvements behind Claude Code Security are available to anyone with API access.

If your team is evaluating these capabilities, the limited research preview is the right place to start, with clearly defined data handling rules, audit logging, and success criteria agreed up front.

Source link

Advertisement
Continue Reading
Click to comment

You must be logged in to post a comment Login

Leave a Reply

Tech

Samsung and AMD deepen AI memory pact with potential foundry deal on the table

Published

on


The agreement positions Samsung as a key supplier for AMD’s next-generation AI products. Under the terms, Samsung will provide its forthcoming high-bandwidth memory, HBM4, for AMD’s Instinct MI455X accelerators – processors designed specifically for AI workloads.
Read Entire Article
Source link

Continue Reading

Tech

Your inbox is someone else’s business model. It doesn’t have to be

Published

on

There is a moment, usually around the third eerily accurate ad for something you only mentioned in an email, when you start to wonder what exactly your inbox knows about you. The answer, it turns out, is everything. And the companies running the most popular free email services in the world are not keeping that information to themselves.

This article contains affiliate links. If you make a purchase through these links, we may earn a commission at no extra cost to you.

The real cost of free email

When Gmail launched in 2004, offering a gigabyte of free storage felt almost absurd. Two decades later, the bargain looks rather different. Google processes roughly 1.8 billion Gmail accounts worldwide, and the service remains free because users are not the customer. Advertisers are.

Every message that lands in a free inbox is parsed, categorised, and fed into a profile that determines which ads follow you across the web. The content of your emails, the receipts, the travel confirmations, the medical appointment reminders, all of it contributes to an advertising profile that you never agreed to build and cannot fully delete. This is not a conspiracy theory. It is the documented business model of the largest email providers on the planet.

Advertisement

For years, most people accepted this trade-off because the alternatives were either expensive, clunky, or both. That is no longer the case.

Advertisement

What private email actually looks like in 2026

The private email market has matured considerably. Services now exist that offer the speed, polish, and reliability of Gmail without the data harvesting. The model is simple: you pay a modest subscription, and in return, your provider has no reason to touch your data because the subscription is the product, not your attention.

Fastmail is one of the more interesting players in this space, and it is worth understanding why. Founded in 1999 in Melbourne, Australia, the company has been running independently for over 25 years. It predates Gmail by half a decade. While most of the tech industry spent the 2010s chasing advertising revenue, Fastmail quietly built a subscription email service focused on something almost quaint: making email work exceptionally well.

The result is a platform that feels noticeably faster than what most people are used to. Full-text search returns results across your entire inbox in milliseconds, not seconds. Keyboard shortcuts cover virtually every action. Server-side filtering rules let you automate sorting and labelling without leaving a client running. These sound like small things until you spend a week with them and realise your previous inbox felt like wading through treacle.

The features that actually matter

It is easy to list features. It is harder to explain why they matter. Here is what stands out after extended use.

Advertisement

Custom domains on every plan. If you own a domain, you can use it with Fastmail immediately. This means your email address is yours permanently, not tied to a provider. If you ever decide to leave, your address comes with you. For freelancers, small businesses, and anyone who treats their email address as professional infrastructure, this is quietly one of the most important features an email service can offer.

Over 600 masked email aliases. Every time you sign up for a new service, you can generate a unique address that forwards to your real inbox. If that address gets sold to spammers or leaked in a data breach, you simply delete it. Fastmail integrates this directly with 1Password, so generating a new alias takes roughly two seconds.

Open standards throughout. Fastmail supports IMAP, SMTP, CalDAV, and CardDAV natively. This means it works with any email client you already use: Apple Mail, Thunderbird, Outlook, or anything else that speaks standard protocols. You are not locked into a proprietary app to access your own email. This level of interoperability is surprisingly rare among private email providers, several of which require bridge applications or restrict you to their own clients entirely.

Calendars and contacts included. Shared calendars, contact management, and file storage are built into every plan. For families and small teams, this replaces the need for a separate productivity suite. The Duo plan covers two users with shared calendars for $8 per month on annual billing, and the Family plan extends to six users at $11 per month.

Advertisement

The privacy question, honestly

Fastmail is transparent about what it does and does not offer. It uses TLS encryption in transit and AES encryption at rest, which is industry standard. It does not offer end-to-end encryption, and the company has been straightforward about why: the usability trade-offs (slower search, limited client support, key management complexity) do not serve most users well.

If your threat model includes protection against server-side access by a state actor, Fastmail is not the right choice. For that, you want a zero-knowledge encrypted provider like ProtonMail. But if your primary concern is getting your inbox out of the advertising ecosystem while keeping a fast, flexible, standards-compliant email experience, Fastmail delivers exactly that. The company does not sell data, does not display ads, does not build user profiles, and publishes regular transparency reports.

This distinction matters because privacy is not binary. Moving from a provider that actively monetises your email content to one that simply does not touch it is a significant upgrade for the vast majority of users, even without end-to-end encryption.

What it costs

Fastmail’s pricing is straightforward. The Individual plan costs $6 per month or $5 on annual billing, and includes 50GB of storage, custom domains, over 600 aliases, calendars, contacts, and full third-party client support. The Duo and Family plans share storage across two or six users respectively, making them some of the most cost-effective private email options for households.

Advertisement

Business plans start at $4 per user per month for Basic (6GB per user), $6 for Standard (60GB), and $10 for Professional (150GB with email retention archiving). Teams can mix and match tiers within a single account, so you are not paying for enterprise-grade archiving on every seat.

There is a 30-day free trial with no credit card required. Full access to every feature, no artificial limitations.

The bigger picture

The argument for private email is not really about email at all. It is about a slow, accumulating shift in how people think about the services they use every day. For two decades, “free” has been the default, and the cost has been invisible. But the cost is there: in the ads that follow you, in the data brokers who know your purchase history, in the quiet erosion of the idea that a conversation can be genuinely private.

Paying $5 per month for an inbox that belongs to you is not a radical act. But it is a meaningful one. And the tools available in 2026 make it easier than it has ever been. Fastmail’s free trial is a reasonable place to start, if only to see what email feels like when nobody is watching.

Advertisement

Source link

Continue Reading

Tech

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

Published

on

from the celebrate-with-some-lemon-pound-cake dept

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 of deliberation.

To recap briefly: deputies raided Afroman’s home in 2022 with guns drawn, found essentially nothing, filed no charges, broke his door (and his gate!), and got caught on his security cameras doing embarrassing things — including one deputy who appeared to cautiously eye a delicious-looking lemon pound cake. Afroman turned the footage into multiple viral music videos. The deputies, upset about being mocked, sued him for $3.9 million claiming defamation and emotional distress. The jury took just a few hours to say: nah.

The best part might be the closing argument from the officers’ attorney, who told the jury:

“Mr. Foreman doesn’t get to wrap himself in the American flag and say you can’t touch me, I can say what I want, no matter how untrue it is, no matter how much pain it causes people, because I have freedom of speech. He can’t do that.”

Afroman’s lawyer quickly responded that he can, in fact, do exactly that. That’s how the First Amendment works. Especially when talking about public officials. And then the jury agreed. This is especially delicious given that Afroman literally wrapped himself in the American flag for the entire trial, showing up each day in that wonderful suit.

Advertisement

Afroman’s own testimony summed up the whole case more concisely than any lawyer could:

“All of this is their fault, and they have the audacity to sue me.”

And through all of this, Afroman never stopped making music mocking these officers — right up to the trial. Here he is calling out Deputy Randy Walters:

And here he is set to the tune of the Battle Hymn of the Republic, reminding everyone that the proof of everything he’s saying is right there on the internet for anyone to see:

Advertisement

So the deputies sued because they were embarrassed by viral music videos, and, in doing so, created a three-day trial that generated a whole new wave of viral content about them, drew national media attention, and ended with a jury telling them they had no case. The Streisand Effect remains undefeated.

As Afroman’s lawyer told the jury in closing, citing NWA’s “Fuck Tha Police” and Richard Pryor’s comedy:

“I’m sorry they feel the way they do, but there’s a certain amount that you have to take as a public official, it’s part of the duties of the job. What chilling effect does that have on the world we live in? You don’t like what a public official does and you make a joke, and you’re dragged into court?”

There’s a serious point underneath all the absurdity. Public officials who raid your home for no good reason, find nothing, and break your stuff don’t get to then use the courts to punish you for talking about it. That’s the whole ballgame on the First Amendment, and the jury understood it perfectly.

Advertisement

Afroman summed it up outside the courtroom saying:

I didn’t win. America won. America still has freedom of speech. It’s still for the people, by the people!

Well said. And I hope the Adams County Sheriff’s Department is looking forward to Afroman’s next release.

Filed Under: 1st amendment, adams county, adams county sheriff’s department, afroman, defamation, free speech

Advertisement

Source link

Continue Reading

Tech

A New Game Turns the H-1B Visa System Into a Surreal Simulation

Published

on

More than half of the nine developers who worked on the game have either obtained a US visa or tried and failed to do so. Most of them are from China, but the team also intentionally recruited talent from other countries in the hopes of incorporating more diverse immigrant perspectives.

“Everybody knows somebody that’s on a visa, but not all of them are vocal about that part of their identity,” says Andrea Saravia Pérez, an immigrant from Colombia who joined the team in February as a narrative designer. “How can we develop a project that’s interactive and shows people this immigration system that a lot of Americans are not familiar with?”

There’s growing interest across the gaming industry in making political games, says Yang. When her team brought H1B.Life to the annual Game Developers Conference in San Francisco last week, she says they received a tremendous amount of interest and support because they are tackling an important societal issue without expecting to make much profit. (The game was supported by a philanthropic organization and the developers also plan to raise additional funding from a future Kickstarter campaign.)

Yang says she has also heard from people in Germany and Australia interested in licensing or adapting the game for different countries. “The whole world is turning right, and life is getting more difficult for all immigrants,” she says.

Advertisement

“If we can just put people in our shoes, I think it can create a very positive impact,” says Saravia Pérez. “As long as players come to have fun and are able to sympathize and understand it a little bit more, I think that we’ve done our job as a team.”

Image may contain Advertisement Poster Text and Purple

Courtesy of Reality Reload

Technicalities Versus Emotions

The H-1B visa program, created in 1990, is one of the most reliable US immigration pathways for white collar workers with college degrees. In recent years, the program issued about 85,000 visas annually, but since there are often more applicants than slots, a lottery system determines who ultimately is chosen. And if you don’t get it, you have to wait an entire year before you try again. Every person who has gone through the process has their own success or failure story to tell, me included.

The team behind H1B.Life started developing the game by interviewing immigrants. So far, Yang says they have talked to over two dozen people about their H-1B journeys and used those interviews to make the game more realistic and accurate. The biggest challenge now is to figure out how to balance explaining complicated immigration rules accurately and ensuring the game is still entertaining.

Source link

Advertisement
Continue Reading

Tech

Bridge of Spirits launches for Switch 2 on March 26

Published

on

The Switch 2 ports keep on coming. This time it’s Kena: Bridge of Spirits, the award-winning 2021 title from Ember Lab. Previously announced for spring 2026, the visually striking title now has an official release date of March 26.

Kena: Bridge of Spirits won Best Independent Game and Best Debut Indie Game at The Game Awards 2021. It’s already available for PS5 / PS4, PC (Steam and Epic) and Xbox Series X/S and One.

You play as Kena, a young spirit guide on a quest to a sacred mountain shrine. Gameplay has a Zelda-like flair. (That could make it a solid next play after Breath of the Wild and Tears of the Kingdom.) Like in Link’s adventures, you’ll find plenty of exploration, puzzles and fast-paced combat. That encompasses whacking bad guys with Kena’s staff, firing arrows and flinging bombs.

Kena: Bridge of Spirits launches for Switch 2 on March 26 in North America, Europe and Asia. Meanwhile, folks in Thailand can get it a day earlier, on March 25. Details about Taiwan will be announced “soon.” You can preorder the game today in North America and Europe and get a taste of its Pixar-esque art style in the trailer below.

Advertisement

Source link

Continue Reading

Tech

This Hermes wireless charger costs four times more than the iPhone you'll charge on it

Published

on

Luxury accessory maker Hermes has announced a new iPhone and Apple Watch charger with its own case for an eye-watering price of $5,150.

Apple Watch with red leather band and orange leather AirPods case on a brown wireless charging pad, resting on a light wooden desk with a metal keyring attached
Hermes’ new charging station can wirelessly power two things at once

If that figure sounds like a lot to pay for a charger, consider this. You could buy eight $599 MacBook Neo laptops and still have money left over for a new Apple Watch instead.
Alternatively, the iPhone 17 Pro Max is the most costly thing you’ll charge with this thing. You could buy four $1,199 256GB iPhone 17 Pro Maxes and still have money left over for a new case. And AppleCare+.
Continue Reading on AppleInsider | Discuss on our Forums

Source link

Continue Reading

Tech

Spotify Exclusive Mode Delivers “Bit Perfect Audio” on Windows Desktop but MacOS and Mobile Users Are Left Waiting

Published

on

Spotify has quietly added a feature that many desktop listeners have been requesting for years. The company’s new Exclusive Mode allows Spotify Premium subscribers using the Windows desktop app to bypass the computer’s normal audio processing and deliver bit perfect playback directly to an external DAC or audio device.

The idea is straightforward. Modern operating systems often resample audio, mix system alerts with music, or apply their own volume processing before the signal ever reaches your DAC. Spotify’s Exclusive Mode takes control of that pipeline and locks the desktop app directly to the output device so the audio stream is passed through without alteration. In practical terms, that means the data sent to your DAC should match the original digital file that Spotify is delivering.

The timing is notable. After years of teasing higher fidelity streaming tiers, Spotify finally launched its long promised Lossless audio option in 2025, putting the service closer to competitors like Apple Music, Amazon Music, and TIDAL that already offer CD quality streaming. Exclusive Mode does not increase the bitrate of Spotify’s stream, but it removes one more layer of processing that could potentially alter the signal on the way out of your computer.

For listeners using a desktop setup with an external DAC, powered speakers, or a headphone amplifier, the feature gives Spotify something it has historically lacked compared with more audiophile focused platforms: a way to deliver the stream without the operating system getting in the middle of it.

Advertisement
Spotify App Lossless Setup
How to Enable Spotify Lossless

What is Bit Perfect Audio?

Bit perfect audio refers to digital playback where the data leaving a music player is identical to the original digital file, with no changes introduced by the operating system, software mixer, or audio driver.

When music is played on a typical computer, the audio often passes through a system level mixer before reaching the output device. During that process the operating system may:

  • Resample the audio to match the system’s selected sample rate
  • Mix in other system sounds like notifications or alerts
  • Apply volume scaling or DSP processing

Each of those steps technically alters the original digital data stream. Bit perfect playback avoids that entirely.

In an exclusive playback mode, the music application takes direct control of the audio output device. The operating system mixer is bypassed, system sounds are blocked, and the audio stream is sent to the DAC at its native sample rate and bit depth.

For audiophiles using external DACs and dedicated headphone or speaker systems, this ensures the converter receives exactly the same digital information contained in the music file or stream. The DAC then performs the only conversion that matters: turning that digital signal into analog sound.

Advertisement

In short, bit perfect playback does not magically improve a recording. What it does is ensure that nothing in the computer changes the signal before it reaches your DAC.

How to Enable Spotify Exclusive Mode

spotify-enable-exclusive-mode-desktop

Getting started with Spotify’s Exclusive Mode takes only a few seconds inside the desktop app. Open Spotify, go to Settings, and scroll down to the Playback section. Under Audio Output, select your preferred device from the dropdown menu. Once your DAC, interface, or audio device is selected, toggle Exclusive Mode to On.

Advertisement. Scroll to continue reading.

When enabled, Spotify takes direct control of that audio device and bypasses the operating system’s standard audio mixer. The result is a direct signal path from the Spotify desktop app to your DAC, avoiding the resampling, mixing, and volume adjustments that can occur when the operating system manages audio output.

There is one important trade off. With Exclusive Mode active, Spotify has sole control of the selected output device. System alerts, browser audio, video calls, and other applications will not be heard through that device while music is playing. If you need to hear other audio sources, those applications will have to use a different output device.

Advertisement

At launch, Exclusive Mode is available only on the Spotify desktop app for Windows. The feature does not currently apply to mobile playback, and Spotify has indicated that macOS support is expected in a future update.

spotify-lossless-24-bit

The Bottom Line

Spotify’s Exclusive Mode is a useful technical upgrade for desktop listeners, but it doesn’t change the limits of Spotify’s audio quality.

The feature simply bypasses the computer’s audio mixer so the Spotify desktop app can send the stream directly to your DAC without resampling or system sounds interfering. That’s good news for listeners using external DACs, headphone amps, or powered speakers on a computer.

What it does not do is increase resolution. Spotify Lossless remains CD quality at 16 bit 44.1 kHz, and the service still does not offer 24-bit/96 kHz or 24-bit/192 kHz streams like Apple Music, TIDAL, or Qobuz.

So while Exclusive Mode ensures the signal leaving Spotify is cleaner, it does not suddenly make Spotify higher resolution than competing services. It simply ensures you hear the stream exactly as Spotify delivers it.

Advertisement

For more information: community.spotify.com

Source link

Advertisement
Continue Reading

Tech

He started S’pore’s 1st rage room. Now, he sells grass for dogs to pee on.

Published

on

Poopee’s indoor grass patches are even drawing franchise enquiries from across Southeast Asia

What do you do when your dog refuses to use a pee pad, your schedule won’t allow constant trips downstairs, and your carpet is slowly turning into a wreck day after day?

For 32-year-old Royce Tan, the answer was simple: start a company called Poopee—and bring a patch of real grass straight into the living room as an indoor potty solution.

It all started with his own struggles at home.

Royce’s dogs were grass-trained, but as they aged, daily trips downstairs became increasingly difficult. Plastic pee pads simply didn’t make sense to them, leaving Royce with the familiar frustration of accidents on the carpet.

Advertisement

That’s when the idea clicked: if dogs instinctively understand grass, why not bring it indoors? “The behaviour problem would solve itself naturally,” he said.

After researching existing options, Royce realised there wasn’t a well-designed real grass solution available locally. And that gap in the market became the starting point for the Eco-Turf, Poopee’s signature product.

Entrepreneurship isn’t new to him

Image Credit: @supersalami_ via Instagram/ Sonia Yeo

Royce is no stranger to entrepreneurship.

Before Poopee, he founded The Fragment Room in 2017, Singapore’s first rage room, where people could smash objects to relieve stress. “It was exciting, chaotic, and honestly a lot of fun,” he recalled.

Yet after a few years, the novelty began to fade, and Royce realised he wasn’t “waking up excited” about the business anymore. It no longer felt like something that “reflected who he was or what he wanted to build.”

Advertisement

Eventually, Royce exited the venture sometime in 2019 and took on a full-time job, which he stayed in for about three years.

While it provided financial stability and a fresh perspective on work, he quickly realised that simply clocking in and out every day wasn’t the life he wanted. He needed to build something again—and that drive ultimately led to Poopee.

Image Credit: @poopee, @bossdedoberman via Instagram

In 2023, Royce self-funded the launch of Poopee, bringing his indoor grass solution to the market.

He did not disclose the exact investment; however, he described it as modest, adding that it required “careful spending and a lot of experimentation.”

Keeping the grass alive

According to Royce, most of the initial capital went into “sourcing grass, developing packaging, logistics testing, and building the website.”

Advertisement

The early stages also involved a lot of trial and error.

Working with real grass meant dealing with a living product. He had to consider drainage, airflow, and how to keep it alive and healthy throughout packaging, transport, and delivery.

Image Credit: @poopee via Instagram

Unlike most products, grass cannot sit in a warehouse for long periods. “It has to be inspected, packed, and delivered within a short timeframe,” said Royce, adding that it took considerable effort to balance freshness, reliability, and cost.

Even after delivery, external factors in customers’ homes—such as sunlight and airflow—can affect the grass. To address this, Royce stays in close contact with customers via WhatsApp, troubleshooting issues and finding solutions directly with them.

“With product businesses, the real learning only begins once customers start using the product,” he said. It’s a very hands-on process, and even today, the team is still refining the product.

Advertisement

Poopee’s Eco-Turf has become a necessity for several dog owners

Poopee’s pop-ups at Singapore Dog Festival (left) and Crane (right)./ Image Credit: @sgdogfest, @poopee via Instagram

It seems Poopee’s Eco-Turf has been well-received by dog owners, Royce shared, although he didn’t disclose exact take-up rates.

While most sales happen online, Poopee also hosts pop-ups at fairs and events to reach new customers and give people a chance to experience the product in person.

Some owners have even told him, “You can’t ever close down because we would be lost without you,” highlighting just how much Eco-Turf has become a part of their daily routine.

Image Credit: @poopee via Instagram

Part of the product’s appeal is how simple it is to use: the grass naturally absorbs urine, so customers don’t have to worry about accidents. Solid waste still needs to be cleaned up, though, and watering the turf every one to two days keeps it healthy.

Owners will know it’s time to replace the patch when dogs start peeing at the edges—a sign it’s saturated with their scent, and they’re looking for a fresh spot.

Typically, each patch lasts between one and four weeks, depending on how frequently it’s used.

Advertisement

That’s why Poopee offers a subscription model, with deliveries weekly, bi-weekly, or monthly. Subscriptions make it easier for owners to maintain a fresh patch and allow Royce to plan production.

Because grass cannot be grown instantly, our nursery needs advance notice of how much to grow. Subscription demand helps us forecast how much grass will be needed ahead of time.

Royce Tan, founder of Poopee

Image Credit: @poopee, @maxfromsingapore via Instagram

Beyond positive reviews from customers, what has surprised Royce even more is that Poopee’s Eco-Turf has been recommended by several veterinarians—particularly for older dogs or pets recovering from surgery who struggle with mobility. They can benefit greatly from the Eco-Turf’s easy-to-access, indoor setup.

That said, Royce occasionally meets owners who say their dogs “hate grass.” In many cases, it simply means the dog was never exposed to it early on. “Dogs adapt to the environments we create for them, so part of what we do is gently reintroduce that natural behaviour,” he explained.

He recommends a few simple strategies for these cases: place the Eco-Turf in the spot where the dog usually relieves itself and gradually move it to the desired location, or use a small amount of pee residue under the turf to attract them with a familiar scent. Positive reinforcement—treats or praise when the dog uses the Eco-Turf—also helps encourage the behaviour.

Advertisement

Beyond just grass patches

Royce Tan setting up Poopee’s pop-up booth at Good Pet Fair 2024./ Image Credit: @poopee via Instagram

Running The Fragment Room taught Royce the value of building a brand people actually connect with.

Even though Poopee solves a functional problem, I never wanted it to feel like a boring utility product. I wanted it to feel like a lifestyle brand for dog owners.

Royce Tan, founder of Poopee

That’s why the team puts thought into shaping Poopee’s brand personality, and expanding into merchandise is part of that.

Image Credit: @supersalami_, @poopee via Instagram

Beyond Eco-Turf and its complementary products—such as deodorising sprays—Poopee now offers items like dog carriers, graphic tees, stickers, and even cooling neck wraps for dogs.

All of the merchandise is designed in-house, with some pieces developed through collaborations. For instance, the brand has worked with local streetwear label Koterie on several of its designs.

Making everyday life with dogs more “natural, comfortable & well designed”

Image Credit: @poopee via Instagram

For now, Royce still handles most aspects of the business himself, supported by a small group of part-timers and freelancers who assist with packing, logistics, and marketing when needed.

The team recently moved from a small warehouse setup into a proper office space—a shift that has improved both working conditions and operations, allowing for more organised packing and better quality control.

Advertisement

While operations remain lean, Royce sees clear demand for products like the Eco-Turf.

“Singapore has a large number of high-rise homes, and many pet owners have busy schedules,” he shared. “Solutions that make pet care easier while still respecting a dog’s natural behaviour are becoming increasingly important.”

As the brand grows, Royce plans to expand beyond just the Eco-turf and build a whole ecosystem of products for pet owners. “The idea is to create a cohesive world around the brand for people who love animals.”

Interest in Poopee is also spreading beyond Singapore. According to Royce, the brand has received franchise enquiries from markets like Malaysia, Thailand, and the Philippines, with discussions still ongoing as the team explores potential regional expansion.

Advertisement

Ultimately, Royce hopes to make everyday life with dogs more natural, comfortable, and well-designed.

If Poopee can become something that dog owners trust and rely on, something that genuinely improves the quality of life for both pets and their owners, then we’ve done what we set out to do.

Royce Tan, founder of Poopee

  • Find out more about Poopee here.
  • Read other articles we’ve written on Singaporean businesses here

Featured Image Credit: @supersalami_, @poopee via Instagram

Advertisement

Source link

Continue Reading

Tech

Portland cybersecurity startup Eclypsium raises $25M to secure AI infrastructure

Published

on

Eclypsium co-founders Yuriy Bulygin, left, and Alex Bazhaniuk. (Eclypsium Photo)

Eclypsium, a Portland-based cybersecurity startup, raised $25 million in new funding to expand its hardware and firmware security platform.

The round was led by PEAK6 Strategic Capital, with participation from Ten Eleven Ventures and an undisclosed bank. Existing investors Andreessen Horowitz, Madrona, Qualcomm Ventures, and others also participated.

Founded in 2017 by former security engineers at Intel, Eclypsium serves private and public sector customers. It scans the hardware and firmware of laptops, servers, network devices, and AI infrastructure for vulnerabilities that sit below the operating system.

The company plans to use the funds to expand further into AI infrastructure and a growing array of edge devices — including GPU servers, NVIDIA BlueField-based appliances, SASE and SD-WAN edge devices, 5G equipment, and CCTV cameras.

The company is led by CEO Yuriy Bulygin and CTO Alex Bazhaniuk and is ranked No. 115 on the GeekWire 200, our list of top tech startups in the Pacific Northwest.

Advertisement

Eclypsium raised $45 million in equity and debt as part of a Series C round last January. Total funding to date is $110 million.

Source link

Advertisement
Continue Reading

Tech

Closing AI learning gaps between leaders and employees

Published

on

Despite heavy investment in AI tools, many organizations are struggling to realize its promised productivity gains. The issue is not the technology itself, but a widening gap between leadership ambition and employee capability. This gap

Ciara Harrington

Chief People Officer at Skillsoft.

Advertisement

Source link

Continue Reading

Trending

Copyright © 2025