Antony Garrett Lisi argues that as machines become sentient, they will compete for resources in a Darwinian manner, but not necessarily in a way harmful to humans. He likens future AI entities to corporations, which are already autonomous entities that operate within legal frameworks. Lisi suggests that as computers become more powerful, they will increasingly take over decision-making roles in corporations, benefiting humanity by providing advanced goods and services. However, he warns that superintelligent machines operated by governments could be dangerous, as governments have the power to create laws and engage in harmful actions for survival. Lisi introduces the concept of Roko’s Basilisk, a hypothetical situation where a future superintelligent entity punishes those who did not help bring it into existence. He concludes that while corporate-run AIs are likely to be beneficial, government-operated AIs could pose significant risks