In the last few years, not many studies on the use of robot-assisted therapy to recover hand function in acute stroke patients have been carried out. This randomized-controlled observer trial is aimed at evaluating the effects of intensive robot-assisted hand therapy compared with intensive occupational therapy in the early recovery phases after stroke with a 3-month follow-up. Twenty acute stroke patients at their first-ever stroke were enrolled and randomized into two groups. The experimental treatment was performed using the Amadeo Robotic System. Control treatment, instead, was carried out using occupational therapy executed by a trained physiotherapist. All participants received 20 sessions of treatment for 4 consecutive weeks (5 days/week). The following clinical scales, Fugl-Meyer Scale (FM), Medical Research Council Scale for Muscle Strength (hand flexor and extensor muscles) (MRC), Motricity Index (MI) and modified Ashworth Scale for wrist and hand muscles (MAS), were performed at baseline (T0), after 20 sessions (end of treatment) (T1) and at the 3-month follow-up (T2). The Barthel Index was assessed only at T0 and T1. Evidence of a significant improvement was shown by the Friedman test for the FM [experimental group (EG): P=0.0039, control group (CG): P<0.0001], Box and Block Test (EG: P=0.0185, CG: P=0.0086), MI (EG: P<0.0001, CG: P=0.0303) and MRC (EG: P<0.0001, CG: P=0.001) scales. These results provide further support to the generalized therapeutic impact of intensive robot-assisted treatment on hand recovery functions in individuals with acute stroke. The robotic rehabilitation treatment may contribute toward the recovery of hand motor function in acute stroke patients. The positive results obtained through the safe and reliable robotic rehabilitation treatment reinforce the recommendation to extend it to a larger clinical practice.