WASHINGTON – In late October, researchers at North Carolina State University alerted Google to a security flaw that could let scam artists send phony text messages to Android phones — a practice called “smishing” that can ensnare consumers in fraud.
Google’s security officials replied in minutes, confirming the flaw and promising to correct it. Within days they had incorporated a fix into the latest version of the Android operating system, Jelly Bean 4.2, and made available a security update for earlier versions.
But for most Android phones, the fix never arrived. For many, it never will.
That is because it is not clear which company — Google, the smartphone makers or the wireless carriers who sell them — bears the ultimate responsibility for the costly process of getting security updates to Android devices. Fixes to known security flaws can take many months to reach individual smartphones, if they arrive at all.
The problem, say security experts, has contributed to making the world’s most popular mobile operating system more vulnerable than rivals to hackers, scam artists and a growing universe of malicious software.
Breaches remain more common on traditional computers than on smartphones, which have been engineered to include security features not found on desktop or laptop machines, experts say.
But outdated software can undermine such protections. If there was a major outbreak of malicious software, the fractured nature of the system for delivering updates could dramatically slow efforts to protect information carried on Android phones — including documents, passwords, contact lists, pictures, videos, location data and credit card numbers.
The risks are particularly serious for businesses and government agencies, whose increasingly popular bring-your-own device policies have created new potential portals for espionage aimed at secure computer systems.
“You have potentially millions of Androids making their way into the work space, accessing confidential documents,” said Christopher Soghoian, a former Federal Trade Commission technology expert who now works for the American Civil Liberties Union. “It’s like a really dry forest, and it’s just waiting for a match.”
Google engineers designed Android to resist hackers and have continually improved it. The company also has worked to purge malicious software from its app store, called Play, minimizing the risk from one possible route of infection.
“We’ve built the system from day one to deal with this kind of world,” said Hiroshi Lockheimer, vice president of Android engineering. “The health of the Android ecosystem is really important to us.”
Yet while each new generation of Android delivers improvements that close off newly discovered avenues of attack, the company has struggled to get updated software to smartphones already in the hands of consumers.
The latest version of Android — the one with the smishing fix — is used by just 1.2 percent of the more than 500 million Android devices worldwide, according to data compiled by Google. The company says it also released a security patch that could repair the flaw in earlier versions of Android, but neither Google nor the wireless carriers could say how many current phones received the patch.
Ars Technica, a news site covering the technology industry, analyzed the update schedules for dozens of the most popular Android smartphones in December and found that most had received only two updates since consumers bought them, sometimes years earlier.
Apple’s iPhone, the leading competitor to Android smartphones, gets operating system updates several times a year. A similar update schedule is common to desktop and laptop operating systems and other software, with updates happening automatically — often with users not even knowing it.
What is different about the Android line of smartphones is that there are dozens of devices made by various manufacturers, such as Samsung, LG and HTC, that tailor the software and its updates to their own specifications. Then wireless carriers, such as Verizon Wireless, AT&T and Sprint, make their own changes and test each update before sending it to consumers over their wireless networks.
The overall process typically takes months and happens far less frequently than recommended by security experts, who call the diffusion of responsibility among several companies “fragmentation.” Blame, too, is spread widely, although it often focuses on the carriers as the most important choke point.
“Supporting five releases of phones is a cost they absolutely don’t want to incur,” said Dmitri Alperovitch, chief technology officer for CrowdStrike, a security company.
Wireless carriers say they seek to release updates promptly, but they acknowledge that the process generally takes months.
“When more than one company is involved in delivering the final product, as is the case with the Android environment, any improvements in the security update process must include all entities involved,” said Ed Amoroso, chief security officer for AT&T. “We all have a collective interest for a fast and consumer-friendly update process, and we intend to coordinate with other providers to see if we can engineer a better solution than the one we have now.”
Verizon Wireless, the largest wireless carrier, and Samsung, the largest Android device maker, both declined to answer detailed questions and said they deliver updates as quickly as possible. Sprint declined numerous interview requests, referring queries to Google.
But security experts say Google by itself has little power to get faster updates to phones. It founded the Android Update Alliance in 2011, along with carriers and device makers, but the initiative has produced little so far.
Last year, Google bought Motorola Mobility, a leading manufacturer of mobile devices, which may eventually lead to faster updates for that company’s products. Google’s record of updating software on its own line of phones and tablets, called Nexus and produced in conjunction with other manufacturers, is better than when phone makers simply adopt the Android system, which Google makes and distributes for free.
The extent of risk to smartphones is a subject of intense debate among security experts. Alperovitch, the expert from Crowdstrike, said most consumers for now face little danger so long as they buy apps through Google’s store and don’t patronize the growing number of third-party stores, which have become popular in China and beyond.
Other experts say the risk is real and growing for all Android users. McAfee, the antivirus company, says it has documented an explosion in the amount of malicious software designed to target the operating system, which runs on three out of four new smartphones worldwide. Some malicious software steals personal information, while others can initiate phony charges that can appear — and often are not detected — on the cellphone bills of consumers.
Trend Micro, another security company, has reported on the spread of Android-based “botnets,” which could allow remote users to take control of thousands or even millions of devices at a time.
For those looking to hack into smartphones, there are many potential entrances: browsers, text messages, emails, cell signals, Wi-Fi signals, Bluetooth connections and, for the latest smartphones, Near Field Communication radios. Some powerful spying software, typically used by governments, allows hackers to switch on cameras or microphones, to watch or listen to smartphone users.
“Now they can hack your life, your physical life, not just your cyberlife,” said Tom Kellermann, a Trend Micro vice president and member of President Barack Obama’s Commission on Cybersecurity.
Such intrusions are difficult and time-consuming, making them unlikely for ordinary users. But security experts warn such tactics could be used against the most valuable targets, such as business executives or senior government officials, especially if they are running outdated software.
“It’s essentially the weak link in the chain,” said Pat Calhoun, a senior vice president at McAfee, a maker of security software. “The cybercriminals have determined that if they want to get into the enterprise, the best way is through the mobile device.”
The smishing vulnerability — so named because it was a version of “SMS phishing,” meaning it sought to trick users into clicking on a malicious link in a phony text message — was not nearly that serious, nor was there evidence that it had yet spread widely. Xuxian Jiang, the computer science professor who reported the flaw to Google, said he has heard numerous reports of smishing attacks in China but few in the United States.
Yet a serious, widespread outbreak could move much faster than the companies involved in updating Android phones are prepared to react, experts say, potentially causing serious breaches for consumers and expenses for carriers that may need to replace phones compromised by malicious programs.
“They can sweep the world in a few hours,” said Kevin Mahaffey, chief technology officer for Lookout, a mobile security firm. “Thankfully that hasn’t happened on mobile yet. But I do see this as a potentially billion-dollar problem.”
Google officials say they can act fast when faced with the most serious breaches, as they did in 2011 when a hacking incident enabled the Iranian government to monitor as many as 300,000 Iranians using Gmail, the free email service provided by Google.
All major computer browsers received swift updates, fixing the problem. Google, meanwhile, updated Android to prevent similar issues in the future and delivered the repair to smartphones with unusual speed.
“There’s a lot of moving pieces to the update process, so we evangelize,” said Lockheimer, the Android engineering official.
But that incident was the last that prompted such aggressive action, he said. And even now, almost 1½ years later, Google says while the repair reached most Android phones, they don’t know how many remain vulnerable.