Previous estimates of vaccine effectiveness (VE) against asymptomatic influenza virus infection based on seroconversion have varied widely and may be biased. We estimated 2022-2023 influenza VE against illness and asymptomatic infection in a prospective cohort. In the HEROES-RECOVER cohort, adults at increased occupational risk of influenza exposure across 7 US sites provided weekly symptom reports and nasal swabs for reverse transcription-polymerase chain reaction (RT-PCR) influenza testing. Laboratory-confirmed influenza virus infections were classified as symptomatic (≥1 symptom) or asymptomatic during the week of testing. Participants reported demographic information and vaccination through surveys; most sites verified vaccination through medical record and immunization registry review. Person-time was calculated as days from the site-specific influenza season start (September-October 2022) through date of infection, study withdrawal, or season end (May 2023). We compared influenza incidence among vaccinated versus unvaccinated participants overall, by symptom status, and by influenza A subtype, using Cox proportional hazards regression adjusted for site and occupation. We estimated VE as (1 - adjusted hazard ratio) × 100%. In total, 269 of 3785 (7.1%) participants had laboratory-confirmed influenza, including 263 (98%) influenza A virus infections and 201 (75%) symptomatic illnesses. Incidence of laboratory-confirmed influenza illness among vaccinated versus unvaccinated participants was 23.7 and 33.2 episodes per 100 000 person-days, respectively (VE: 38%; 95% CI: 15%-55%). Incidence of asymptomatic influenza virus infection was 8.0 versus 11.6 per 100 000 (VE: 13%; 95% CI: -47%, 49%). Vaccination reduced incidence of symptomatic but not asymptomatic influenza virus infection, suggesting that influenza vaccination attenuates progression from infection to illness.