Astrophysicists Rethink What It Means for a Planet to be “Habitable” & ChatGPT Helps with the Tomato Harvest (Patreon)

Astrophysicists Rethink What It Means for a Planet to be “Habitable”

The term “habitable” has traditionally been used to describe planets that could have liquid water on their surface. But this definition now seems overly simplistic given the diversity of exoplanets scientists have discovered. A new paper by researchers from France and the U.S. suggests a better method to characterize habitability and inhabitation of exoplanets based on their atmospheric properties and biosignatures. The method uses a combination of spectroscopy and machine learning to classify planets into four categories: uninhabitable, habitable, inhabited and inhabited with technological life. There’s currently only one planet in the latter category, but the authors argue that this method could help us to better identify what observations to focus on. Paper here. More here.

New Precision Test of Standard Model Gets Green Light

The U.S. Department of Energy has approved Jefferson Lab’s plans for a new experiment that will measure the weak charge of the electron. Dubbed the MOLLER experiment, it will use electrons from a particle accelerator at the lab to precisely measure how electrons scatter off other electrons. The acronym stands for “Measurement of a Lepton-Lepton Electroweak Reaction,” and in case you’re wondering why they say “lepton” when they mean “electron”, it’s because the process is known as Møller scattering, after the Danish physicist Christian Møller. Besides, MOEEER sounds dumb. 

Measuring the weak charge of the electron is a relatively inexpensive precision test for the predictions of the standard model. If everything goes as planned, the experiment could start collecting data by late 2025, with first results in early 2027. Press release here.

ChatGPT Helps with the Tomato Harvest

While everyone is griping about how AI is going to take over the world, researchers from EPFL and TU Delft have used ChatGPT to design a robotic gripper for harvesting tomatoes. The researchers engaged in a dialogue with ChatGPT to define the robot’s specifications, features, materials, and control code. They then fabricated and tested the robot in the real world, followed by troubleshooting and some modifications, but overall it worked well. According to the authors, their study demonstrates the potential of ChatGPT as a creative partner for human engineers and inventors. Paper here, press release here.