What We're Reading

The scientific literature is riddled with bad charts and graphs, leading to misunderstanding and worse. Avoiding design missteps can improve understanding of research.

Microscopy has been a trade-off until now: the bigger the sample, the lower the resolution. But picking out cellular detail in mouse brains and more is becoming increasingly possible.

Understanding how artificial intelligence makes its decisions is often very difficult to do. Google hopes to solve that problem by feeding counterfactual data to AI systems to figure out what the systems are paying attention to.

In the briny lakes in Africa’s Rift Valley and the cold, dry soil of Antarctica’s Shackleton Glacier Valley, life reaches its limits: Researchers have pinpointed conditions too hostile for even the toughest microbes, contradicting the current wisdom that life is really everywhere.

The Environmental Protection Agency must desist from a course that could harm both human and planetary health, argues Nature's editorial board.

Scientists face an ethical dilemma over what to do with their ‘human bycatch.’

"Superintelligence” is a flawed concept and shouldn’t inform our policy decisions, argues Melanie Mitchell, a professor of computer science at Portland State University.

An experiment that partially revived slaughterhouse pig brains raises questions about the precise end point of life, writes neuroscientist Christof Koch.

Debate highlights broader concerns about tools that use the results of direct-to-consumer genetic testing.

In an experiment to boost transparency, the preprint server will allow select journals and independent peer-review services to publicly post evaluations of its papers should the authors make the request.