Cyber solidarity on the chopping block.

CISA Layoffs threaten U.S. cyber coordination with states, businesses, and foreign partners. Google issues its second emergency Chrome update in a week, and puts Privacy Sandbox out of its misery. OpenAI’s new browser proves vulnerable to indirect prompt injection. SpaceX disables Starlink devices used by scam compounds. Reddit sues alleged data scrapers. Blue Cross Blue Shield of Montana suffers a data breach. A new Android infostealer abuses termux to exfiltrate data. Iran’s MuddyWater deploys a wide-ranging middle east espionage campaign. We’re joined by Lauren Zabierek and Camille Stewart Gloster discussing the next evolution of #ShareTheMicInCyber. When customer service fails, try human resources.

Today is Thursday October 23rd 2025. I’m Dave Bittner. And this is your CyberWire Intel Briefing.

CISA Layoffs threaten U.S. cyber coordination with states, businesses, and foreign partners.

The East Wing of the White House isn’t the only thing the Trump administration is taking a wrecking ball to. The president has effectively shuttered the Cybersecurity and Infrastructure Security Agency’s (CISA) Stakeholder Engagement Division (SED), a key unit responsible for coordinating cybersecurity improvements with state, local, private, and international partners. Sweeping layoffs in mid-October cut nearly all 95 staff, leaving only the Sector Management unit intact. The move eliminates three vital offices—Council Management, Strategic Relations, and International Affairs—disrupting CISA’s partnerships with critical infrastructure operators, academic institutions, and foreign governments.

Experts warn the cuts will erode trust, reduce situational awareness, and weaken collaboration mechanisms essential for defending sectors such as healthcare, energy, and telecommunications. Former White House cybersecurity adviser Michael Daniel said the downsizing risks leaving CISA “blind to certain threats and trends.” Industry and government officials described the cuts as “dangerous,” potentially isolating the U.S. from global cyber allies and depriving defenders of shared intelligence and expertise. CISA said the restructuring was meant to “realign the agency’s mission.”

Google issues its second emergency Chrome update in a week, and puts Privacy Sandbox out of its misery. 

Google has issued its second emergency Chrome update in a week, patching a high-severity flaw in the browser’s V8 JavaScript engine. Tracked as CVE-2025-12036, the vulnerability was discovered by Google’s AI-driven research project, Big Sleep. Details remain undisclosed until most users update. The fix appears in version 141.0.7390.122/.123 across Windows, Mac, Linux, and Android. Users can trigger the update manually via Chrome’s “About Google Chrome” settings page to ensure immediate protection.

Elsewhere, Google has officially ended its long-delayed Privacy Sandbox project, abandoning plans to replace third-party cookies with privacy-focused ad technologies. The company will retire ten remaining Sandbox APIs. Citing complexity, poor adoption, and regulatory pressure, Google will instead maintain cookies alongside limited privacy tools like CHIPS and FedCM. The move restores short-term ad stability but highlights the industry’s continued lack of viable, privacy-safe alternatives.

OpenAI’s new browser proves vulnerable to indirect prompt injection. 

OpenAI’s new Atlas browser has been found vulnerable to indirect prompt injection—a technique where malicious instructions hidden in web content trick AI agents into unintended actions. Brave Software’s report this week confirmed the flaw as a systemic problem across AI-powered browsers like Perplexity’s Comet and Fellou. While Atlas resisted some tests, researchers still demonstrated successful injections using Google Docs, prompting Atlas to output manipulated text.

OpenAI acknowledged the risk, calling prompt injection an “unsolved security problem” despite red-teaming, safety guardrails, and detection systems. Experts, including AI researcher Johann Rehberger, warn that the threat undermines data confidentiality and integrity, with no perfect fix yet. Rehberger urged stronger downstream security and human oversight. OpenAI maintains Atlas remains experimental and is refining protections for safer AI browsing.

SpaceX disables Starlink devices used by scam compounds. 

SpaceX has disabled more than 2,000 Starlink satellite devices used by scam compounds in Myanmar after mounting pressure from politicians and anti-trafficking advocates. Lauren Dreyer, SpaceX’s vice president of business operations, said the company proactively shut down over 2,500 kits near suspected scam centers and is cooperating with global law enforcement. The move follows reports that Starlink had enabled internet access for cybercrime operations near the Thai border, despite previous government crackdowns.

Thai and U.S. officials, including Senator Maggie Hassan, had urged Elon Musk to prevent Starlink’s use in human trafficking and large-scale fraud. Myanmar authorities recently seized dozens of Starlink devices in a raid that detained over 2,000 people at a major scam complex. While some operations have been disrupted, reports suggest new compounds continue to emerge despite ongoing enforcement.

Reddit sues alleged data scrapers. 

Reddit has filed a lawsuit against four companies—SerpApi, Oxylabs, AWMProxy, and Perplexity—accusing them of illegally scraping Reddit content via Google search results and selling it to AI developers like OpenAI and Meta. Reddit seeks a permanent injunction, damages, and a ban on using scraped data. The company argues that AI firms are fueling a “data laundering economy” by exploiting its user-generated content without compensation. While SerpApi and Perplexity deny wrongdoing, Reddit says it spent millions building anti-scraping defenses and even trapped Perplexity with a hidden test post to prove its case. The lawsuit underscores rising tensions between data owners and AI companies as content-rich platforms move to license data rather than give it away. Reddit already has paid data deals with Google and OpenAI.

Blue Cross Blue Shield of Montana suffers a data breach. 

Blue Cross Blue Shield of Montana is under investigation after a data breach exposed personal and medical information for up to 462,000 residents. The breach, spanning November 2024 to March 2025, may have compromised names, addresses, and billing data. State Auditor James Brown called the incident “deeply disturbing” and launched an immediate probe, criticizing the insurer for failing to notify customers or provide credit monitoring. Officials urge Montanans to monitor benefits statements and report suspicious activity.

A new Android infostealer abuses termux to exfiltrate data. 

SANS researcher Xavier Mertens has uncovered an infostealer that uses Termux on Android to run Python, harvest contacts, messages, location, app data and banking-related files, and exfiltrate them via Telegram.

The sample (SHA256: 7576cdb8…) scored 0/64 on VirusTotal and includes Vietnamese comments, the researcher reports. The malware calls Termux utilities and scans mapped storage paths for Facebook, WhatsApp, media, and banking filenames. It installs a persistent backdoor script that periodically logs location data and sends information using a Telegram bot token.

The finding shows Android can be a vector for infostealers traditionally focused on Windows, elevating risks to confidentiality and integrity of user data.

Iran’s MuddyWater deploys a wide-ranging middle east espionage campaign. 

Iranian state-sponsored hacking group MuddyWater (also known as Static Kitten, Mercury, and Seedworm) has targeted over 100 government entities across the Middle East and North Africa using version 4 of its Phoenix backdoor, according to a new Group-IB report. The campaign began August 19 with phishing emails sent from a compromised account accessed via NordVPN. Attached Word documents contained malicious macros that deployed the FakeUpdate loader, which decrypted and installed Phoenix v4.

The new version features enhanced persistence through COM objects and supports commands for file upload, download, and shell execution. Researchers also found an infostealer targeting Chrome, Edge, Opera, and Brave browsers for credentials. MuddyWater additionally leveraged the PDQ and Action1 RMM tools for remote control. Most victims were embassies, consulates, and foreign ministries, signaling continued Iranian cyber-espionage operations.

When customer service fails, try human resources. 

In a move equal parts Mission: Impossible and petty genius, a TikTok user named @kerneldump shared how he discovered a company’s data breach — and when no one at customer service would listen, he applied for a job there. After breezing through multiple interviews, he finally sat down with the Chief Information Security Officer, who cheerily asked what drew him to the company. His answer? The dark web alert showing his data — from their site — for sale. He wasn’t there for the paycheck; he was there to deliver the breach notice in person. Then he declined the job. Cybersecurity might be an arms race, but sometimes it only takes one determined applicant to remind a company that “defense in depth” should include answering your emails.

And that’s the CyberWire.

For links to all of today’s stories, check out our Daily Briefing at the cyberwire dot com.

We’d love to know what you think of this podcast. Your feedback ensures we deliver the insights that keep you a step ahead in the rapidly changing world of cybersecurity. If you like the show, please share a rating and review in your podcast app. Please also fill out the survey in the show notes or send an email to cyberwire@n2k.com

We’re proud that N2K CyberWire is part of the daily routine of the most influential leaders and operators in the public and private sector, from the Fortune 500 to many of the world’s preeminent intelligence and law enforcement agencies.

N2K helps cybersecurity professionals and organizations grow, learn, and stay ahead. We’re the nexus for discovering the people, tech, and ideas shaping the industry.  Learn how at n2k.com.

N2K’s senior producer is Alice Carruth. Our producer is Liz Stokes. We’re mixed by Elliott Peltzman and Tré Hester, with original music by Elliott Peltzman. Our executive producer is Jennifer Eiben. Peter Kilpe is our publisher, and I’m Dave Bittner. Thanks for listening.