With the rise of social networks, powerful actors have increasingly attempted to manipulate public opinion in their favor using various methods, including the deployment of artificial accounts such as bots and trolls. The influence of these artificial entities poses substantial risks to fundamental freedoms, including thought, expression, and decision-making, while also creating significant social, political, cultural, and economic challenges for society. To mitigate this influence, it is essential to identify and analyze the factors that contribute to the effectiveness of artificial entities. Despite significant advancements in recent years in the detection, modeling, and analyzing of the effects of artificial entities, less attention has been paid to the accurate identification of the factors influencing their role in the formation of public opinion. This gap persists, particularly in the context of simultaneously considering both consensus-building and polarization-inducing stimuli. In this study, we try modeling artificial entities in social networks and conduct a series of experiments under various conditions to examine the factors that influence their success. To this end, we also introduce new criteria for assessing the formation of public opinion and the effectiveness of artificial entities, addressing the current lack of suitable metrics in this domain. The study demonstrates that attachment strategy, intelligence level, and lifespan of artificial entities play critical roles in shaping public opinions. We show that even a small presence of artificial entities can lead to consensus, underscoring the importance of their characteristics in shaping public discourse. By understanding the factors that influence the success of artificial entities in manipulating public opinion, policymakers, social network platforms, governing organizations, and individuals can effectively counteract their influences.