Hello Nature readers, would you like to get this Briefing in your inbox free every day? Sign up here.
The FCC (large circle, dashed outline) would be built close to its predecessor at CERN, the Large Hadron Collider (small circle).Credit: CERN
There are “no technical or scientific showstoppers” that stand in the way of a new US$17-billion, 91-kilometre supercollider underneath the French and Swiss countryside, according to a feasibility study by CERN, Europe’s particle-physics laboratory. It’s unclear what the Future Circular Collider might find: theory gives no clear steer on what might be discovered at such high energies, and physicists are at odds over whether this is an argument against, or in favour, of the new machine.
Nature | 5 min read
A general-purpose AI system can be turned into a chemistry specialist with only a little tweaking. Researchers added information already gathered about chemical compounds or materials to the training data of GPT-3, an early iteration of the large language model that powers ChatGPT. This allowed the system to predict properties of similar materials and reaction yields as well as, or better than, more specialized algorithms. “This greatly reduces the barrier for other chemists to benefit from machine learning in their domains,” says chemical engineer Andrew White.
Nature | 5 min read
Reference: Nature Machine Intelligence paper
Scientists who think of their research as being high-risk spend more time applying for funding. A survey asked more than 4,000 US-based academics to rate how risky they feel their research is, and how they think their peers would perceive its riskiness. Fewer than 10% of respondents had a high average risk score — 8 or more out of 10 — and this correlated strongly with more time on grant applications. “If you are doing more uncertain things, one could imagine that it’s harder to get funding,” says economist and study co-author Kyle Myers. “You have to convince peers that it’s a good idea.”
Nature | 6 min read
Reference: arXiv preprint (not peer reviewed)
The European Commission says it wants to cut net greenhouse-gas emissions by 90% compared with 1990 levels by 2040. The goal, which is not yet legally binding, is part of the bloc’s commitment to achieve ‘climate neutrality’ by 2050. Some researchers worry that the ambitious target risks relying heavily on largely unproven carbon removal technology — rather than prioritizing cutting fossil fuels. The election in some EU countries of governments less committed to climate action might also make the goal difficult to achieve.
Nature | 5 min read
Features & opinion
Table of Contents
The partially submerged volcano that forms the Greek island of Santorini experienced a gargantuan blast more than half a million years ago. An expedition that drilled into the sea floor discovered signs of an eruption bigger than one in around 1600 BC that might have contributed to the decline of the Minoan civilization on the nearby island of Crete. Researchers also found that a blast in AD 726 was much bigger than previously thought. “The 726 eruption has always been used as a worst-case scenario” for a modern eruption at Santorini, says expedition volcanologist Timothy Druitt. “What’s interesting is, the worst-case scenario has just increased in magnitude quite a lot.”
Nature | 7 min read
‘Allostatic load’ is the stress response triggered by chronic adverse situations, resulting in cumulative wear and tear. It describes a familiar experience for women, particularly those of colour, in the scientific workforce, writes neuroscientist Jean King. “On an individual level, what is helpful is finding your people,” she says. “The need to build communities might seem to add to cognitive load, but I think the opposite is true: seeing other minoritized people doing what we do gives us great strength.”
Nature | 5 min read
There is still much work to do to make science more collaborative, transparent, accessible, equitable and inclusive, finds a report from the cultural organization UNESCO — and embracing open science can help achieve the UN’s Sustainable Development Goals, humanity’s best attempt to map a pathway towards a better future. “The UNESCO report shines a much-needed light on some promising developments in open science,” says a Nature editorial. “The challenge will be how to accumulate individual examples of good practice into something similar to a critical mass.”
Nature | 5 min read
Infographic of the week
Networks of self-propelled, ‘active’ fluids — such as suspensions of bacteria — follow rules similar to the brain-bending game sudoku. In the classic version of the game, you have to fill a grid of numbers without repeating them. In this case, when the active fluid hits a three-way junction, the net flow through one channel must be zero to conserve fluid volume across the node. Predicting active flows through a network is like playing a sudoku-like game in which the object is to place the three numbers +1, 0 and −1 at every network node using the fewest zeros possible (a). Add more nodes and things get more complicated: wide channels give rise to segregated looping flows, whereas loops are more often nested in networks of narrow channels (b). The reason for this lies in how vortexes form in a channel that has zero net flow (c). (Nature News & Views | 7 min read, Nature paywall)
Reference: Nature Physics paper (free to read)