Enlarge / Wolf Point Creek is likely the most-well-studied glacier-fed stream in the world. (credit: Elizabeth via Flickr (CC BY-NC 2.0) )

Pushing off from the dock on a boat called the Capelin , Sandy Milner’s small team of scientists heads north, navigating through patchy fog past a behemoth cruise ship. As the Capelin slows to motor through humpback whale feeding grounds, distant plumes of their exhalations rise from the surface on this calm July morning. Dozens of sea otters dot the water. Lolling on backs, some with babes in arms, they turn their heads curiously as the boat speeds by. Seabirds and seals speckle floating icebergs in this calm stretch of Alaska’s Glacier Bay.

Some two hours later, the craft reaches a rocky beach where Wolf Point Creek meets the sea. The creek is a relatively new feature on the landscape: Land at its mouth first became ice-free in the 1940s due to the melting and retreat of a glacier. It took shape through the 1970s, fed by a mountain lake that slowly formed as an isolated chunk of glacier ice slowly melted. Wolf Point Creek is special because almost its entire life span — from the first, sparse trickles melting out under the ice edge to a mature stream ecosystem teeming with aquatic life, from tiny midge larvae to small fish, and with willows and alder weaving along its edges — is known in intimate detail, its history painstakingly documented.

Milner, a stream ecologist at the University of Birmingham in the UK, has returned almost annually to this spot since the 1970s to catalog how life — particularly aquatic invertebrates — has arrived, thrived and changed over time. He was here to observe meager midges in 1977 and to spot a hundred prospecting pink salmon in 1989. A decade later, his team cataloged 10,000 of the fish spawning in Wolf Point Creek.

Enlarge / Recycling is sorted at the Sims Municipal Recycling Facility in New York City in 2015. (credit: Spencer Platt/Getty Images)

T o jumpstart a paltry market for recycled plastic, governments across the globe are pushing companies to include recycled materials in their products. Last year, the United Kingdom introduced a tax on manufacturers that produce or import plastic packaging containing less than 30 percent recycled plastic. In 2024, New Jersey will begin enforcing similar rules, albeit with lower targets. California now requires that beverage containers be made of 15 percent recycled materials, and Washington will enact a similar requirement later this year. The European Commission , Canada , and Mexico are all considering comparable moves.

Currently, most plastic products are derived from freshly extracted fossil fuels, including crude oil and natural gas. Incorporating some recycled plastic could reduce emissions, and shrink pollution in waterways and landfills, experts say. But collecting, sorting, pulverizing, and melting post-consumer plastics for reuse is expensive. The new laws will potentially help recyclers find buyers for what would otherwise become waste.

But regulators may need a better way to verify that the new laws are working. While companies can enlist a third-party to certify their use of recycled content, most certifiers take a bird’s-eye view, tracking the materials across a range of products and factories. As a result, an item with a “recycled content” label might be completely devoid of recycled content.

Enlarge (credit: Walter Zerla via Getty Images)

Say you are a maker of computer graphics cards, under pressure from investors questioning your green credentials. You know what to do. You email your various departments, asking them to tally up their carbon emissions and the energy they consume. Simple enough. You write a report pledging a more sustainable future, in which your trucks are electrified and solar panels adorn your offices.

Good start, your investors say. But what about the mines that produced the tantalum or palladium in your transistors? Or the silicon wafers that arrived via a lengthy supply chain? And what of when your product is shipped to customers, who install it in a laptop or run it 24/7 inside a data center to train an AI model like GPT-4 (or 5) ? Eventually it will be discarded as trash or recycled. Chase down every ton of carbon and the emissions a company creates are many times times higher than it first seemed.

Enlarge / Even hermit crabs have individual patterns of behavior — personalities, if you like. When scientists ignore the effects of such differences, they may produce research that’s flawed. (credit: NurPhoto via Getty Images )

Several years ago, Christian Rutz started to wonder whether he was giving his crows enough credit. Rutz, a biologist at the University of St. Andrews in Scotland, and his team were capturing wild New Caledonian crows and challenging them with puzzles made from natural materials before releasing them again. In one test , birds faced a log drilled with holes that contained hidden food, and could get the food out by bending a plant stem into a hook. If a bird didn’t try within 90 minutes, the researchers removed it from the dataset.

But, Rutz says, he soon began to realize he was not, in fact, studying the skills of New Caledonian crows. He was studying the skills of only a subset of New Caledonian crows that quickly approached a weird log they’d never seen before—maybe because they were especially brave, or reckless.

The team changed its protocol. They began giving the more hesitant birds an extra day or two to get used to their surroundings, then trying the puzzle again. “It turns out that many of these retested birds suddenly start engaging,” Rutz says. “They just needed a little bit of extra time.”

Enlarge (credit: Edward Olive/Getty Images)

ChatGPT might well be the most famous, and potentially valuable , algorithm of the moment, but the artificial intelligence techniques used by OpenAI to provide its smarts are neither unique nor secret. Competing projects and open source clones may soon make ChatGPT-style bots available for anyone to copy and reuse.

Stability AI , a startup that has already developed and open-sourced advanced image-generation technology, is working on an open competitor to ChatGPT. “We are a few months from release,” says Emad Mostaque, Stability’s CEO. A number of competing startups, including Anthropic , Cohere , and AI21 , are working on proprietary chatbots similar to OpenAI’s bot.

Enlarge (credit: Anadolu via Getty Images )

It has been more than a decade since Jeff Bezos excitedly sketched out his vision for Alexa on a whiteboard at Amazon’s headquarters. His voice assistant would help do all manner of tasks, such as shop online, control gadgets, or even read kids a bedtime story.

But the Amazon founder’s grand vision of a new computing platform controlled by voice has fallen short. As hype in the tech world turns feverishly to generative AI as the “next big thing,” the moment has caused many to ask hard questions of the previous “next big thing”—the much-lauded voice assistants from Amazon, Google, Apple, Microsoft, and others.

A “grow grow grow” culture described by one former Amazon Alexa marketing executive has now shifted to a more intense focus on how the device can help the e-commerce giant make money.