Skip to Content

The internet is taking over a person’s life for Halloween. And you can be a part of it.

An MIT experiment is handing a single person’s free will to the crowd to test how the digital hive mind works.
October 31, 2018
BEEMEBeeMe

Tonight, you can take control of someone else. Well, you and a few hundred people can. The MIT Media Lab will be performing a Halloween experiment to put the internet hive mind to the test. And it will require one mystery actor to surrender free will to that crowd. 

The test, called BeeMe, will start in an undisclosed location at 11 p.m. US eastern time tonight. Users will then be faced with the task of defeating a rogue AI by guiding an actor’s actions. How they do that—and whatever else they get up to along the way—is completely up to the internet. While a general plot has been put in place, the creators didn’t want to limit the interactions too much.

The MIT team created an interface where people can see and hear what the actor is doing and seeing, as well as suggest the next actions to take. The researchers estimate that the experiment, which will move east to west across MIT’s campus, will take about two hours. That is, if people stay on track.

“One of the features of intelligent behavior is the ability to make long-term plans,” says Niccolo Pescetelli, one of the researchers leading the project. “Right now we don’t have any evidence crowds can do that. The planning is lacking, since there is no central control. We are pushing the limits in that direction.”

But it’s not just about completing the route as quickly as possible. For the researchers, success will be seeing how the participants push the bounds of the challenge. During tests using 10 to 15 people, the crowd approached strangers, random bystanders tried to help the actor with directions across campus, and the group entered into conversations that took an awkwardly long time. Pescetelli says that this time, as soon as one command is completed, another will be loaded up ready for the actor to execute.

So what’s the point, apart from a bit of spooky Halloween fun?

Tests of digital crowdsourcing are nothing new. In 2014, thousands of people used the game-streaming service Twitch to play the video game Pokémon Red en masse using crowdsourced commands. And past projects by the Scalable Cooperation group at the Media Lab have explored similar ideas, such as how groups can write horror stories collaboratively.

But the experiment tonight will push internet cooperation to the next level, requiring quick suggestions and voting that will guide what is happening in the real world. The hope is that it will give insight into how real crowds work together, and provide clues as to how such systems can be used in more serious situations. “When there is a time-critical task like an election or a rally, things can be unpredictable,” says Manuel Cebrian,   the group’s leader. “In a way, we wanted to reflect on that.”

If you want to become a part of the internet hive mind, you can tune into the stream tonight at 11.

 

Keep Reading

Most Popular

AV2.0 autonomous vehicles adapt to unknown road conditions concept
AV2.0 autonomous vehicles adapt to unknown road conditions concept

The big new idea for making self-driving cars that can go anywhere

The mainstream approach to driverless cars is slow and difficult. These startups think going all-in on AI will get there faster.

biomass with Charm mobile unit in background
biomass with Charm mobile unit in background

Inside Charm Industrial’s big bet on corn stalks for carbon removal

The startup used plant matter and bio-oil to sequester thousands of tons of carbon. The question now is how reliable, scalable, and economical this approach will prove.

images created by Google Imagen
images created by Google Imagen

The dark secret behind those cute AI-generated animal images

Google Brain has revealed its own image-making AI, called Imagen. But don't expect to see anything that isn't wholesome.

AGI is just chatter for now concept
AGI is just chatter for now concept

The hype around DeepMind’s new AI model misses what’s actually cool about it

Some worry that the chatter about these tools is doing the whole field a disservice.

Stay connected

Illustration by Rose WongIllustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.