A closed-loop insulin system, often labelled the "artificial pancreas" (AP), consists of an insulin pump, a continuous glucose monitor, and an interface coordinating between them to regulate insulin dosage based on glucose levels. Primarily designed for managing type 1 diabetes, this system has demonstrated significant benefits in previous studies. Yet, despite these advantages, certain challenges persist. Semaglutide, utilized in treating type 2 diabetes and obesity, is a once-weekly injectable medication that elevates levels of a gastrointestinal hormone known as Glucagon-Like Peptide-1 (GLP-1). This hormone alters gastric emptying, inhibits glucagon release, and reduces appetite. While not officially sanctioned for type 1 diabetes treatment in North America, studies have explored its efficacy as an adjunctive therapy alongside insulin, yielding favorable outcomes in blood glucose regulation. Comparable drugs like liraglutide and exenatide have been employed in type 1 diabetes treatment as well, albeit with less pronounced glucose-regulating effects compared to semaglutide, even in type 2 diabetes. The goal of this 50-week randomized placebo-controlled crossover 2x4 factorial designed trial is to assess whether commercial automated insulin delivery (AID) systems using rapid-acting insulin with adjunct weekly injections of semaglutide (at the maximally tolerated dose) can replace carbohydrate counting with simple meal announcements (SMA) without degrading glucose control.

The main questions this study aims to answer are: * Can weekly injections of semaglutide at the maximum tolerated dose in individuals with T1D on closed-loop therapy with SMA and rapid-acting insulin result in a non-inferior time spent in target range (3.9-10 mmol/L) compared to weekly placebo injections on closed-loop system with full carbohydrate counting. * Can weekly injections of semaglutide at the maximum tolerated dose, in combination with ultra-rapid actin insulin (Lyumjev); 1. Eliminate carbohydrate counting and any meal announcement (i.e fully closed-loop) in people with T1D on closed-loop therapy without degrading glucose control. 2. Be more effective in substituting carbohydrate counting with SMA in people with T1D on closed-loop therapy compared with traditional rapid-acting insulin. Participants will be asked to undergo two subsequent blinded drug interventions; one with semaglutide and the other with placebo. Both interventions include 4 meal strategies each with a 3-week duration; full carbohydrate counting with rapid-acting insulin, SMA with rapid-acting insulin, SMA with Lyumjev and fully closed-loop system with Lyumjev.