After a series of security and privacy mishaps, Zoom has taken further steps to allay fears of its privacy and security, the company announced in a blog post yesterday. To start with, the company has created a Chief Information Security Officers Council (CISO Council) to advise it, with CISOs from HSBC and NTT Data, among others. The company is also tapping a subset of the council to advise Zoom CEO Eric S. Yuan personally, with security experts from Netflix, VMware and Uber.
The company will also rope in cybersecurity expert and Facebook’s former chief security officer Alex Stamos of Stanford University as an external advisor; Stamos has actively tweeted about Zoom’s privacy issues.
Zoom is going to need to demonstrate more transparency, including putting a security face to all of these responses. A documented 30 day security plan that includes a feature freeze, several professional pen-tests and rolling out coordinated disclosure policies would be smart.
— Alex Stamos (@alexstamos) April 1, 2020
Zoom was banned from government use in Taiwan on 7 April over security concerns, and from schools in New York City one day earlier. On April 1, the company announced a feature freeze (as Stamos suggested above), as scrutiny on it built up as the company grew in size by a factor of 20. Here are Zoom’s security issues up until this point, in this year alone.
- On April 8, Motherboard reported that several hackers are showing interest in so-called “zero-day” exploits, which are vulnerabilities that are not disclosed to anyone. The report says that the hackers are trying to sell these exploits to the highest bidder, which means there could be risks in the app that cannot be patched until the attacks actually happen. The report does not name any of the hackers.
- On April 3, the Washington Post reported that several video call recordings were left in the open web, without any password protection required to access recordings. The videos included classes and therapy sessions.
- On April 2, the New York Times reported a feature where even anonymous participants in a call could siphon off LinkedIn data about other participants without their knowledge. Zoom later removed this feature.
- On April 1, researchers discovered a vulnerability where Windows users could have their operating system’s login password stolen with a malicious link sent on chat. CEO Yuan said that this issue has been patched.
- On March 31, The Intercept reported how Zoom was misleadingly claiming that calls were end-to-end encrypted, when they were not. The company later changed the language to say that “Your client connection is encrypted”.
- On March 31, security researcher Felix Seele discovered that Zoom’s macOS installation package was working around Apple’s requirements for installing apps by just extracting a compressed archive directly into Apple computers’ Applications folder, a tactic commonly employed by malware, and not legitimate software companies. CEO Yuan responded on Twitter and a fix was rolled out two days later.
- On March 30, New York’s Attorney General Letitia James sent a letter to Zoom questioning its data privacy and security practices, the New York Times reported.
- On March 26, a Motherboard investigation found that Zoom was sending user data to Facebook even if users were not logged in via the social media platform, or had an account there. Zoom updated the app to remove the Facebook Software Development Kit that was causing this data leak.
One major concern has been “zoombombing”, where people enter meetings that they are not meant to be included in; this has been an especially serious issue for Zoom, since the security settings by default on calls allowed anyone with a valid link to join the call. On April 4, the company announced that it would turn the Waiting Rooms feature on by default for all users, which would require hosts to manually approve each new caller.