Could Microsoft solve the scareware problem?

This morning I read the following article: Microsoft can help kill fake antivirus threat. And interesting approach. The proposal is that we could white-list all the legitimate security software within the OS in order to make it harder to trick the user. Well, would this work? I am not so sure:

• First of all, what is Security Software and how do you find out? All the the security vendors can play by the rules and make sure it is detectable. But sacreware (fake anti-malware software) will probably not – or will for sure not. So, what is the difference between any legitimate application, any application which interacts with the desktop and presents a GUI vs. scareware? Scareware just show scary windows and makes you install their software – which is typically malware.
• The base technology is in Windows but it would have to be applied to security software only.
• What is legitimate security software? There are obvious ones like Symantec’s, McAfee’s, TrendMicros’, F-Secure’s, Microsoft's solutions. That’s easy. But I am sure (just an experience from the past) that there will be a pretty big gray zone which makes it very hard to decide and who decides then – us?
• Last but not least, let’s talk about the regulators. Do they (and does the market) really want us to take this decision and “certify” anti-malware solutions? This would come with a price – and reading the comments in the article below, this is one of the issues.

To me, the problem is wider spread than “just” fake anti-malware solutions. I understand that this is a problem – definitely and I understand that the thoughts of white-listing security software is attractive. But the problem is malware in general and how the criminals trick the user into installing something they do not want. This leads back to the question of the trusted stack which we address in our End to End Trust vision. To me, that’s the only approach which can be successful

Roger