US Edition
Rocket Scientists Hooked Up ChatGPT to the Controls of a Spaceship, and the Results Were Not What You Might Expect
3 Articles
3 Articles
Rocket Scientists Hooked Up ChatGPT to the Controls of a Spaceship, and the Results Were Not What You Might Expect
AI Kirk Despite being stumped by simple children's games and uncontrollably hallucinating, AI models could perform surprisingly well when put in charge of navigating space inside a simulated spacecraft, researchers have found. As Live Science reports, a team of scientists from MIT and the Universidad Politécnica de Madrid, Spain, instructed OpenAI's blockbuster AI chatbot ChatGPT to "operate as an autonomous agent controlling a pursuit spacecraf…
In a special competition, engineers at the Massachusetts Institute of Technology used the large language model behind ChatGPT to task it with controlling a spaceship. It was all just a game, of course, but it was very useful.
ChatGPT has demonstrated excellent performance during a recent spaceship pilot test under the Kerbal Space Program Differential Game Challenge program, the first evaluation of the genre for large language models (LLM). After converting the textual responses of the models into functional code for the autonomous piloting of a hypothetical ship, that of OpenAI has won second place among the best performers. LLM is an important advance in the field …
Coverage Details
Bias Distribution
- 100% of the sources lean Left
To view factuality data please Upgrade to Premium
