Interestingly enough, Asia has more connected users than the western world ~825M, but there penetration is only 21%, and in North America it is 77%. So, if you simply bring Asia up to %50 penetration, you would have already added close to 1 billion users (50% of today). That would be 3 billion, since 6b - 3b = 3 billion left to grow, 1 billion growth may in fact be low, conservative, actually.
Now, lets look at devices, this article claims 5 billion devices already:
5b devices with 2b users, an average of 2.5 devices per user.
Now, let's assume that all the current users only double their device usage (I would expect more). That would make 5b x 2 = 10b devices, close enough certainly to not make his claim ludicrous. Add 50% more Asian users to the equation at the same device rate, and boom 15b devices! Not far fetched at all. Again, this seems conservative to me, and it arrives at his estimate.
I have a feeling that users will actually do more than doubling their devices. Internet devices have typically been in the $500-$1000+ range. They are approaching the $100 on the low end now, this is a massive difference and will likely easily more than double the current # of devices per user and could greatly impact penetration also.
Given some facts, show me logic that would make his prediction seem ludicrous.
> Right, but the original statement said 1 billion new users to go with those 15 billion connected devices, not 5 billion total users:
>> Over the next few years, there will be one billion new internet-connected users and 15 billion connected devices, Fisher said.
No, while he could mean what you think he means, it would be a small leap to assume that. I don't read it that way. I do not think that he associated the 1 billion new users with those 15 billion connected devices. I believe those were two independent predictions, related only in context, not a strict causation. I could be wrong about my interpretation, but clearly it is a valid one given that sentence, and (as you would likely agree) it certainly makes more sense given the numbers.
After reading this article, I suspect that his prediction was by 2015. Note, that the Ericsson CEO thinks that there will be 50b connected devices by 2020: