Not OK, Google: Chromium voice extension pulled after spying concerns

Google has removed an extension from Chromium, the open source sibling to the Chrome browser, after accusations that the extension was installed surreptitiously and subsequently eavesdropped on Chromium users.

The issue first came to light in late May when a bug was filed in the Debian bug tracker. Chromium version 43 was seen downloading a binary extension from Google, and there was neither any ability to prevent this download, nor any source code available for the extension. The extension, called "Chrome Hotword," was found to be responsible for providing the browser's "OK, Google" functionality. Although off by default, both Chrome and Chromium, when set to use Google as their default search engine, can permanently listen to the microphone and respond instantly to voice queries, with "OK Google" used as the trigger keyword.

Concern about the nature and purpose of the extension was compounded by the way the browser did and didn't disclose the extension's existence. The list of extensions visible at chrome://extensions/ doesn't include Hotword. Conversely, Hotword's own status page, chrome://voicesearch/ said that by default the extension was enabled and had access to the microphone.

Read 10 remaining paragraphs | Comments

Fraud Alert Issued on Business Email Compromise Scam

Original release date: June 24, 2015

The Financial Services Information Sharing and Analysis Center (FS-ISAC) and federal law enforcement agencies have released a joint alert warning companies of a sophisticated wire payment scam referred to as business email compromise (BEC). Scammers use fraudulent information to trick companies into directing financial transactions into accounts scammers control.  

Users and administrators are encouraged to review the BEC Joint Report for details and refer to the US-CERT Tip ST04-014 for information on social engineering and phishing attacks.


This product is provided subject to this Notification and this Privacy & Use policy.


OPM director on security issues: We’re trying very hard

With the total number of people affected by the data breach at the Office of Personnel Management now estimated to be as many as 18 million, OPM Director Katherine Archuleta has mounted a public relations counter-attack, defending the agency's efforts to improve security during her tenure and crediting those efforts with finding the malware at the heart of the breach in the first place. But the news of the exposure has caused a wave of fear and distrust among federal employees—with some who work in the intelligence community now concerned for their families' safety.

Archuleta defended her tenure before a Senate hearing on June 23. "I'm as angry as you are that this is happening," she said in a message to federal employees and retirees during her testimony. "I am dedicated to ensuring that OPM does everything in its power to protect the federal workforce, and to ensure that our systems will have the best cyber security posture the government can provide.” And she insisted that no one at OPM was to blame for the breaches, saying, "If there is anyone to blame, it is the perpetrators."

Today, OPM e-mailed an eight-page document outlining OPM's "Actions to Strengthen Cybersecurity and Protect Critical IT Systems" to members of the media. In the document, OPM officials asserted, "Upon Director Archuleta’s arrival, OPM engaged in an end-to-end review of its IT systems and processes. Based on that review, the agency developed a Strategic Plan for Information Technology to guide its efforts to protect its legacy systems to the maximum extent possible as it replaced them with more modern and secure systems. This plan laid out a multi-phase strategy to bolster security through realignment of professional staff, adherence to relevant laws, policies and best practices, and investments in modern tools."

Read 14 remaining paragraphs | Comments

Patch early, patch often: Adobe pushes emergency fix for active 0-day

Yet again, Adobe has released a new patch to fix a critical vulnerability that "could potentially allow an attacker to take control of the affected system," according to the company.

Adobe acknowledged that the flaw (CVE-2015-3113) is "being actively exploited in the wild via limited, targeted attacks." Known affected systems run Internet Explorer for Windows 7 and below and Firefox on Windows XP, according to the patch details. Adobe says the following software can potentially be impacted:

  • Adobe Flash Player 18.0.0.161 and earlier versions for Windows and Macintosh
  • Adobe Flash Player Extended Support Release version 13.0.0.292 and earlier 13.x versions for Windows and Macintosh
  • Adobe Flash Player 11.2.202.466 and earlier 11.x versions for Linux

The company recommends updating to the latest version of Flash to avoid the risk of exploitation, but at this point users should take a hard look at how necessary Flash is to their daily Internet use. In 2015 alone, we've seen Adobe issue multiple emergency Flash updates to patch critical vulnerabilities under active attack—including three such instances in the first five weeks of the year. The situation has gotten so grim that security reporter Brian Krebs recently experimented with a month without having the Flash Player installed at all. "The result? I hardly missed it at all," Krebs writes.

This newest flaw was uncovered through the help of FireEye security researchers. A Singapore-based FireEye team discovered the vulnerability in June by detecting a phishing campaign exploiting CVE-2015-3113. "The attackers’ e-mails included links to compromised Web servers that served either benign content or a malicious Adobe Flash Player file that exploits CVE-2015-3113," FireEye writes.

FireEye identified APT3, a China-based group also known as UPS, as responsible for these attacks (see more on the group in FireEye's report on Operation Clandestine Fox). APT3 has previously introduced other browser-based zero-day attacks against Internet Explorer and Firefox. FireEye notes APT3's tactics are difficult to monitor given there's little overlap between campaigns, and the group typically moves quickly ("After successfully exploiting a target host, this group will quickly dump credentials, move laterally to additional hosts, and install custom backdoors," the new report states). According to the security researchers, APT3 has implemented these phishing schemes against companies in aerospace and defense, engineering, telecommunications, and transportation this year.

FireEye's report on CVE-2015-3113 offers much greater detail than Adobe's patch notes. For instance, the typical phishing e-mails were spam-like offers for refurbished iMacs:

"Save between $200-450 by purchasing an Apple Certified Refurbished iMac through this link. Refurbished iMacs come with the same 1-year extendable warranty as new iMacs. Supplies are limited, but update frequently.

Don't hesitate . . .>Go to Sale"

FireEye also broke down where unfortunate targets were directed after clicking such URLs—a compromised server hosting JavaScript profiling scripts. "Once a target host was profiled, victims downloaded a malicious Adobe Flash Player SWF file and an FLV file," FireEye reports. "This ultimately resulted in a custom backdoor known as SHOTPUT, detected by FireEye as Backdoor.APT.CookieCutter, being delivered to the victim’s system. The payload is obscured using xor encoding and appended to a valid GIF file."

Read on Ars Technica | Comments