Google admits Hangouts doesn’t use end-to-end encryption, opening the door for government wiretaps

Google admits Hangouts doesn't use end-to-end encryption, opening the door for government wiretaps

If you’re really worried the government may be keeping tabs on your conversations, then you’d best avoid Hangouts.

According to Motherboard, a Google representative confirmed that Hangouts conversations are only encrypted “in transit,” meaning after the message arrives at the intended recipient Google could access it if forced to do so by a government wiretap.

The question arose from a Reddit AMA with two senior members of Google’s public policy and legal team. An ACLU representative pinned them down about encryption, but wasn’t able to get them to detail if all messages were encrypted from end-to-end.

Richard Salgado, Google’s director for law enforcement and information security, and David Lieber, the senior privacy policy counsel, would only confirm the in-transit encryption. Salgado reaffirmed the government’s prerogative to order such surveillance: “There are legal authorities that allow the government to wiretap communications.”

In reality, such wiretaps are rare. Google’s transparency report details only seven wiretap orders for nine accounts in the first half of 2014, the most recent data available because the U.S. government requires a six-month waiting period.

Why this matters: Apple has touted the privacy of iMessage as another advantage to the security conscious over Android. Other messaging platforms, like the Mark Cuban-backed Cyber Dust, also promise secrecy. Google may not see this extra step as necessary until a backlash arises from those who want more privacy from their Hangouts conversations.

Google is Keeping the NSA Out of Your Data, Eric Schmidt Brags

Google is Keeping the NSA Out of Your Data, Eric Schmidt Brags

Google (GOOGL) Chairman Eric Schmidt boasted on Wednesday about how improving the encryption of Google’s products has successfully shut out warrantless surveillance by the NSA and other law enforcement. Schmidt talked about the encryption advances, and how former NSA contractor Edward Snowden’s leaks prompted them, at BoxDev, a yearly developers conference for Box.

“When the Snowden revelations came out, we were very, very upset,” Schmidt told Aaron Levie, CEO of Box. “They never said anything to us. So we embarked upon a program to fully encrypt the information that our customers entrusted to us.”

Encryption makes it very difficult or impossible for information passed electronically to be deciphered, either by the NSA or even by the company doing the encryption. Snowden’s leaks showed how the NSA uses warrantless mass surveillance of metadata, which Schmidt argued went beyond proper use of the Patriot Act. He and other tech company leaders started boosting their encryption to keep the security agencies from being able to read any email or communication without a warrant. Now encryption is not just a Google project, and it appears to be working.

“Apple and others did the same,” Schmidt said. “And we know our program works, because all the people doing the snooping are complaining about it.”

He’s right about that. FBI Director James Comey told Congress that they should ban phone encryption because of how it helps criminals get away with their crimes. The surveillance is party of what the tech and Internet industry wants to see changed in the Patriot Act and why they are hoping it won’t be renewed in its present form.

Google and Yahoo Encrypting Ad Network Connections

Google and Yahoo Encrypting Ad Network Connections

Google and Yahoo in separate announcements said they will individually encrypt ad network connections to reduce bot traffic and other types of ad fraud. The news coincides with the release of Malwarebytes Labs findings last week. Researchers found malvertising in Flash ads involving the DoubleClick ad network.

The two companies have support. The Interactive Advertising Bureau (IAB) continues to push the adoption of HTTPS ads and support encryption. In March, the IAB put out a call for the industry to adopt encryption. The industry trade group said many ad systems support HTTPS, but a member survey suggests that only 80% support the protocol. They called on the entire advertising supply chain to adopt practices, from ad servers and beacons to data partners and brand safety and verification tools.

Google said the majority of mobile, video, and desktop display ads on its Google Display Network, AdMob, and DoubleClick networks will become encrypted by June 30. Search on google.com is encrypted for a vast majority of users and the copany continues to work toward encrypting search ads across its systems.

YouTube ads have been encrypted since the end of last year, along with all searches, Gmail, and Drive. By the end of June, advertisers using AdWords and DoubleClick will serve HTTPS-encrypted display ads to all HTTPS-enabled inventory.

Yahoo VP of Revenue Management and Ad Policy James Deaker describes in a blog post what he calls “perhaps the largest-ever transition to SSL encryption for any publisher with display ads.” Yahoo recently implemented an end-to-end encryption extension for Yahoo Mail,” and strengthening security everywhere else along the advertising supply chain will help to create a safer Internet.

Next week, Yahoo will host a Trust UnConference in San Francisco, bringing together industry experts to discuss how to build safe products.

Another Reason For Ubiquitous Web Encryption: To Neuter China’s “Great Cannon”

Another Reason For Ubiquitous Web Encryption: To Neuter China's "Great Cannon"

China’s web censorship machine, the Great Firewall, has a more offensive brother, researchers have declared today. Called the Great Cannon by Citizen Lab, a research body based at the University of Toronto, it can intercept traffic and manipulate it to do evil things.

In recent distributed denial of service (DDoS) attacks on code repository Github, the Great Cannon was used to redirect traffic intended for Baidu , the equivalent of Google in China, to hit two pages on the target site, including one that provided links to the Chinese-language edition of the New York Times. GreatFire.org, a website dedicated to highlighting Chinese censorship, was hit by a similar attack.

The Great Cannon only intercepts traffic to or from a specific set of targeted addresses, unlike the Great Firewall, which actively examines all traffic on tapped wires going in and out of China. According to Citizen Lab, in the recent DDoS hits, it intercepted traffic going to Baidu, and when it saw a request for certain JavaScript files on a Baidu server, it appeared to either pass the request on “unmolested”, as it did for 98 per cent of connections, or it dropped the request before it reached Baidu and sent a malicious script back to the requesting user, as it did nearly 2 per cent of the time. That malicious script would fire off traffic to the victims’ servers. With so many users redirected to the targets, the internet pipes feeding Github and GreatFire.org were clogged up, taking them offline. It was an effective, if blunderbuss, approach to censoring the targets.

But, as the researchers noted, the Great Cannon could be abused to intercept traffic and insert malware to infect anyone visiting non-encrypted sites within the reach of the attack tool. That could be done, said Citizen Lab, by simply telling the system to manipulate traffic from specific targets, say, all communications coming from Washington DC, rather than going to certain sites, as in the abuse of Baidu visitors. “Since the Great Cannon operates as a full man-in-the-middle, it would also be straightforward to have it intercept unencrypted email to or from a target IP address and undetectably replace any legitimate attachments with malicious payloads, manipulating email sent from China to outside destinations,” Citizen Lab added in its report released today.

The Great Cannon is not too dissimilar to QUANTUM, a system used by the National Security Agency and the UK’s GCHQ, according to the Edward Snowden leaks.  So-called lawful intercept providers, FinFisher and Hacking Team, sell products that appear to do the same too, Citizen Lab noted.

But there’s one simple way to stop the Great Cannon and the NSA from infecting masses of users: encrypt all websites on the internet. The system would not be able to tamper with traffic that is effectively encrypted. The SSL/TLS protocols (which most users commonly use when on HTTPS websites rather than HTTP) drop connections when a “man-in-the-middle” like the Cannon is detected, whilst preventing anyone from peeking at the content of web communications.

There are some significant projects underway designed to bring about ubiquitous web encryption. Just this week, the Linux Foundation announced it would be hosting the Let’s Encrypt project, which seeks to make SSL certificates, which website owners have to own and integrate into their servers to provide HTTPS services, free and easy to acquire. It should be possible to grab these simple and (hopefully) secure certificates from mid-2015, though Josh Aas, executive director at the the Internet Security Research Group (ISRG), which runs Let’s Encrypt, would not say when exactly. It has some serious backers, including Akamai, Cisco, Electronic Frontier Foundation and Mozilla.

It’s unclear whether Let’s Encrypt would provide certificates to Chinese sites. “The default stance is that we want to issue to everyone – but we will have to comply with US laws… our legal team is looking into it.”

“There’s a lot of the web that isn’t encrypted,” added Jim Zemlin, executive director at The Linux Foundation. “We think that’s a big deal for internet security.”

Microsoft Windows also vulnerable to ‘FREAK’ encryption flaw

Microsoft Windows also vulnerable to 'FREAK' encryption flaw

Computers running all supported releases of Microsoft Windows are vulnerable to “FREAK,” a decade-old encryption flaw that leaves device users vulnerable to having their electronic communications intercepted when visiting any of hundreds of thousands of websites, including Whitehouse.gov, NSA.gov and FBI.gov.

The flaw was previously thought to be limited to Apple’s Safari and Google’s Android browsers. But Microsoft warned that the encryption protocols used in Windows — Secure Sockets Layer and its successor Transport Layer Security — were also vulnerable to the flaw.

“Our investigation has verified that the vulnerability could allow an attacker to force the downgrading of the cipher suites used in an SSL/TLS connection on a Windows client system,” Microsoft said in its advisory. “The vulnerability facilitates exploitation of the publicly disclosed FREAK technique, which is an industrywide issue that is not specific to Windows operating systems.”

Microsoft said it will likely address the flaw in its regularly scheduled Patch Tuesday update or with an out-of-cycle patch. In the meantime, Microsoft suggested disabling the RSA export ciphers.

The FREAK (Factoring RSA Export Keys) flaw surfaced a few weeks ago when a group of researchers discovered they could force websites to use intentionally weakened encryption, which they were able to break within a few hours. Once a site’s encryption was cracked, hackers could then steal data such as passwords, and hijack elements on the page.

Researchers said there was no evidence hackers had exploited the vulnerability, which they blamed on a former US policy that banned US companies from exporting the strongest encryption standards available. The restrictions were lifted in the late 1990s, but the weaker standards were already part of software used widely around the world, including Windows and the web browsers.

“The export-grade RSA ciphers are the remains of a 1980s-vintage effort to weaken cryptography so that intelligence agencies would be able to monitor,” Matthew Green, a Johns Hopkins cryptographer who helped investigate the encryption flaw, wrote in a blog post explaining the flaw’s origins and effects. “This was done badly. So badly, that while the policies were ultimately scrapped, they’re still hurting us today.”