A new Atari-playing AI appears to use some underhanded tricks to get high scores when left to its own devices.
What’s new: An AI that learns through a trial-and-error technique called evolution strategies has been pitted against eight Atari games. Its approach gradually mutates the way it tackles tasks, keeping hold of the successful tricks and discarding ones that don’t work.
Any means necessary: But New Scientist notes that when playing the arcade classic Q*bert, the AI developed some unusual winning strategies. It found a software bug that it could exploit to get points, and a trick where carefully planned suicide allowed it to progress through the game.
Why it matters: On one hand, it shows how evolutionary approaches let AI succeed without any human help. But it’s also a reminder that we may need to place limits on which strategies AIs are allowed to use in order to achieve their goals.
This artist is dominating AI-generated art. And he’s not happy about it.
Greg Rutkowski is a more popular prompt than Picasso.
What does GPT-3 “know” about me?
Large language models are trained on troves of personal data hoovered from the internet. So I wanted to know: What does it have on me?
An AI that can design new proteins could help unlock new cures and materials
The machine-learning tool could help researchers discover entirely new proteins not yet known to science.
DeepMind’s new chatbot uses Google searches plus humans to give better answers
The lab trained a chatbot to learn from human feedback and search the internet for information to support its claims.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.