Developer's Account Suspended by OpenAI for Utilizing ChatGPT to Operate a Turret (Automatic Rifle)

A developer on the Reddit forum created a turret equipped with an automatic rifle, driven by ChatGPT's Realtime API. The developer could control the direction and firing of the turret simply through voice commands.

For instance, in a video, the developer gave the command: "ChatGPT, we are being attacked from the front-left and front-right." Subsequently, the AI-driven turret rapidly fired towards the mentioned directions.

It's noteworthy that the turret was primarily operated by voice commands rather than a real-time vision API. After receiving a voice command, ChatGPT would react and coordinate with the turret's control board to perform actions like turning and elevating.

The most impressive aspect was the AI's response speed. The execution of shooting operations could take less than a second after a command was given, or up to 2-3 seconds in slower instances.

However, this developer's actions clearly violated OpenAI's usage agreement, leading OpenAI to identify and suspend the associated account. (How was the video's corresponding account identified? Was it through data related to the conversation?)

This case of a public video leading to an account suspension is just one example. The weaponization of artificial intelligence is inevitable, and it may already be underway, albeit with few public disclosures.