Yes, a lot of laypeople were expecting them to outright prevent infection at much higher rates, but that was never the deal from the start. Consider that the clinical trial protocols only called for testing participants for COVID if they started showing symptoms; the numbers presented were never about preventing infection entirely.
The lower percentages quoted in the article are still pretty fantastic, despite their drop from the initial 90%+ numbers from the clinical trials. Remember that the clinical trials were also done pre-delta variant. It's pretty impressive that they've held up so well against delta.