Many philosophers hold that we should proportion our degree of belief to the strength of the available evidence. It would be bad epistemic practice to believe without evidence. Imagine someone forming the belief that aliens built the pyramids because someone paid them to believe it. This would be irrational partly due to the lack of evidence.
It stands to reason — if we should base our degree of belief on the evidence — that it would be a bad idea to let what we want to believe dictate what we actually believe. Such a practice would quickly lead us to beliefs that are highly implausible and perhaps even dangerous.
So what gives with the famous phrase — seen on a poster in Fox Mulder's office in the X-Files — ‘I want to believe?’ It expresses Fox Mulder's strong desire to find evidence that extraterrestrials visited Earth. This attitude may strike us as epistemically ill advised, to say the least, because believing based on what we want to believe could be dangerous or a waste of time.
There is, however, an epistemic benefit to wanting to believe that something is true — provided that one doesn't let their desire to believe something override the rationality of belief, where the degree of it belief matches the strength of the available evidence.
What's the epistemic benefit? Mulder alludes to it in the first episode of season seven:
You don't want to believe. You're not looking hard enough.
Here Mulder alludes to why wanting to believe can be epistemically beneficial when tempered by the requirement that beliefs are supported by evidence: the acquisition of knowledge can require gathering evidence. One motivation for expending the time, effort, and opportunity costs — if you're gathering evidence, you often cannot do other stuff — to acquire such evidence is wanting to believe the claims the evidence supports.
That doesn't mean that one will find evidence to support what they want to believe. However, it does provide the impetus for seeking out evidence that, if found, would support what one wants to believe. The trick is not allowing what one wants to believe to override the evidence. Easier said than done. And in cases where you do not want to believe, or remain indifferent, there are fewer motivations to seek out evidence for such beliefs.
Perhaps wanting to believe — when tempered by the requirement for sufficient evidence for belief— isn't totally epistemically bad. It may be necessary, sometimes, to motivate seeking out evidence.