Security Experts Uncover More Flaws in Microsoft Apps

IDG News Service (London Bureau) — Security experts have discovered more vulnerabilities in Microsoft Word and other software, although hackers do not appear to be exploiting them yet. The flaws have been reported just as Microsoft releases its latest round of security patches.

Three of the vulnerabilities affect Word 2007, according to the Security Vulnerabilities website. Details are scant at the moment, but two of them appear to allow an attack that can create conditions similar to those caused by a denial-of-service attack, with CPU usage surging to 100 percent, the posting said.

The third vulnerability could allow remote code execution, and the fourth, which concerns the ".hlp" extension for Windows help files, could lead to a heap overflow condition, the posting said.

Three proof-of-concept Word documents plus a malicious ".hlp" file illustrating the vulnerabilities were available for download from at least one website on Wednesday.

Microsoft said it was investigating the reports but was not aware of any attacks as of Wednesday morning.

The discovery of the new vulnerabilities came as Microsoft issued seven fixes for critical flaws on Tuesday. Hackers have often timed the disclosure of new vulnerabilities just after Microsoft's patch day, the second Tuesday of the month, to maximize their time to exploit computers, said Greg Day, a security analyst for McAfee.

"It's becoming a very common trend," he said.

Security researchers have said that as Microsoft fixes problems within its operating systems, hackers are actively hunting for flaws in its Office applications.

When they find one, hackers will send spam with, for example, a malicious Word document attached. Downloading and opening the file could allow a hacker to take control of the machine. Microsoft has warned that people should not open files sent from unknown sources.

April is proving to be a rough month for Microsoft. It issued an emergency patch on April 3 for the animated cursor flaw, which could let a hacker control the machine after a user merely views a malicious website.