Not really sure why this ended up being a Nature paper...that's usually reserved for major insights and findings that break new ground.
All of this is pretty run of the mill virology/immunology that has been known for some time from multitudes of other studies.
Also, if you look at the evolutionary relationship between omicron and precursor strains like delta, you see there's a huge evolutionary leap in omicron. There's absolutely zero surprise that previous infection didn't help much.
See figure 3
www.nature.com
Finally, omicron has fundamentally different biological underpinnings. It's how it was able to evade existing immunity to begin with and become the dominant strain and it's why the newer strains continue to circulate just like other common cold viruses and other human coronaviruses do.
This article is also extremely misleading in frustratingly predictable ways for media outlets, even ones as STEM oriented as Ars Technica.
Sterilizing immunity is extremely uncommon, even in the cases of most current vaccines. Which means the majority of vaccines used and the majority of natural infections don't give 100% immunity forever. Not nearly as ominous as they make it sound, certainly not, IMO, worthy of yearly updated boosters in the population at large.
The primary way this narrative is spun dishonestly, however, is presenting antibody-mediated immunity as THE immunity while giving lip service to the far more important aspects of immunity, which they don't even do in this article despite mentioning its effects.
The only bright spot in the new data was that regardless of what a person was infected with"pre-omicron or omicron"protection from severe, critical, or fatal COVID-19 during reinfection was nearly 100 percent, and that level of protection was sustained for over a year.
This is what's seen with countless other seasonal respiratory viruses - new strains evolve to evade antibodies, which allows infection of a previously immune host...but T cell immunity is not affected nearly as much or at all. So, it's a mild, sometimes subclinical infection with the most risk being present in populations where cellular immunity is waning or lost (ie elderly or people being medically treated with immunosuppressants).
They're describing a transition to a seasonal virus similar to dozens of others.