The article defines what it means by trust in this way.
> Trust, as defined by organizational scholars, is our willingness to be vulnerable to the actions of others because we believe they have good intentions and will behave well toward us.
By the definition, there are many circumstances where rational, selfish individuals can trust each other: when their interests align in a way that communicating honestly and working together is optimal for their own selfish profit.
So, there is room for trust. In addition, the game-theoretic space is large enough and the lack of global information is severe enough that it is not certain in many situations whether betraying trust is indeed the optimal decision in the long run.
Hence in many situations even in the most supposedly rational institutions there is a lot of leeway for its people to decide how the institution should communicate, and its strategy for profit.
On the other hand, the reality is that many institutions aren't rational at all but heavily embody the culture and motivations of its decision makers. You can even think of it rationally like this: letting X person make decisions, regardless of their irrationality, has Y and Z assets, and it is in the company's best interests to secure Y and Z assets by letting X make decisions.
In any case, I think that viewing institutions as rational machines optimizing for profit is hopelessly naive, and likewise for the conclusion.