To be sure, there certainly are many introverts who are perfectly happy on their own with no need to get out and meet people. More power to them! But there are many that crave human connection, even if they happen to have many intellectual interests and for these types of individuals, they would be well served at least carving out some portion of their time to get out of the house with the explicit aim of meeting people. And yes, not every such outing will lead to lifelong friends or meeting your next soulmate, but it's a numbers game.
That's a sense of risk and caution that gets too comfortable for some people to compete with over time. If you don't build yourself better options, all you want to do is sit at home and do the thing that guarantees a reward. Then you get in your car and move about the world in a way that you feel is guaranteed to protect you from conditions, other people, but really is dangerous. You bet only on certainty, and outcomes are predictable, but they're not compatible with not being lonely
But reducing loneliness is just a means to an end. My point is that there exist a lot of rewarding things that you can do alone at home, which may give you a hapiness malus because of the loneliness, but also a happiness bonus because you like the activity.
If a solution to reducing loneliness shall be sustainable, it better increases the happiness or rewardingness overall, too. Otherwise you see loneliness as a problem, but see the alternatives as being the worse options, i.e. by rational choice, the loneliness will not be reduced.
https://www.amazon.com/Yourself-First-Chinese-Nishimura-Hiro...
apparently a lie-flat manual for Chinese and Japanese Gen-Z
Founder of 2chan who reverse took over 4chan in a pattern which should be familiar to you :)
Friendship is a two way contract: you add something to someone's life and they will consider you their friend, they add something to your life, making them your friend.
If you "optimize" for your own and only your own benefit, nobody is going to be your friend.
This is the wrong model:
Sitting (alone) at home and working on program code or reading scientific textbooks does have a reward. Many things for which you go outside of the house or where you interact with other people have a much lower reward. So you rather loose a rather decent local optimum, and if you don't know very well where to look outside for something really good, you get much worse results than if you simply stayed at home and do there what you love.