Lifeboat Foundation

QuEra, a quantum computing startup founded by researchers from Harvard and the Massachusetts Institute of Technology, recently released what may be the most ambitious quantum technology roadmap we’ve seen yet.

The company plans on releasing a quantum computer with 100 logical qubits and 10,000 physical qubits by 2026. It also claims this planned system will demonstrate “practical quantum advantage,” meaning they’d be capable of useful computation feats that classical, binary computers aren’t.

Producing photons one at a time on demand at room temperature is a key requirement for the rollout of a quantum internet—and the practical quantum computers that would undergird that network. The photons can be used as quantum bits (qubits), the quantum equivalent of classical computing’s 0s and 1s. Labs around the world have devised various ways to generate single photons, but they can involve complex engineering techniques such as doped carbon nanotubes or costly cryogenically-cooled conditions. On the other hand, less complicated techniques such as using traditional light sources do not provide the necessary level of control over single-photon emissions required for quantum networks and computers.

Now, researchers from Tokyo University of Science (TUS) and the Okinawa Institute of Science and Technology have collaborated to develop a prototype room temperature single-photon light source using standard materials and methods. The team described the fabrication of the prototype and its results in a recent issue of the journal Physical Review Applied.

“Our single-photon light source … increases the potential to create quantum networks—a quantum internet—that are cost-effective and accessible.” —Kaoru Sanaka, Tokyo University of Science.

Next on the list is a skill called Splash, which uses AI to create a song based on your description. With this one, you can tell Alexa to create any type of song or specify the genre of music you want. After playing a short excerpt, Alexa asks if you want to make any changes, such as adding lyrics. If you like the tune, you can then tell Alexa to send it to your phone for playback.

To use the skill to generate any type of song, say: “Alexa, create a song with Splash Music.” To include a genre, say: “Alexa, open Splash Music,” and you’ll be asked what style you prefer. For this one, I asked Splash to compose a song in the style of jazz. The result certainly wouldn’t win any awards at the next Grammys, but it showed promise.

Finally, the third skill is named Volley Games, an AI-driven spin on the usual 20 questions game. In this one, a friendly AI host challenges you to guess an object by asking the right ‘yes’ or ‘no’ questions. After giving you the category, you’re prompted to ask each question until you hopefully guess the mystery item. Along the way, you can ask for hints if you’re stumped.

Gate-defined superconducting moiré devices offer high tunability for probing the nature of superconducting and correlated insulating states. Here, the authors report the Little–Parks and Aharonov–Bohm effects in a single gate-defined magic-angle twisted bilayer graphene device.

While already deployed for the likes of NASA and ESA for several years, Zeiss’ hologram-generating Multifunctional Smart Glass technology is only now gearing up for mass production. The results could be interesting.

Zeiss’s Multifunctional Smart Glass has been a core and expensive bespoke component of space missions for some years now, the tech having been developed primarily for deployment with the likes of NASA and ESA that can afford it for mission-critical uses. Calling what it results in a hologram is a bit of a cheat — these images don’t float free on their own; instead, they’re generated within a thin, transparent layer sandwiched between glass sheets to which ultra-high-precision optics are attached.

Still, the effect is convincing. The image layer is 92% transparent. Zeiss reckons the holographic functionality can turn any glass surface (windows of buildings, transparent screens, side windows of vehicles, etc) into an on-demand communication screen.

By Viviane Callier

The research conducted by Elena Hassinger, an expert in low-temperature physics working at ct.qmat—Complexity and Topology in Quantum Matter (a joint initiative by two universities in Würzburg and Dresden), has always been synonymous with extreme cold.

In 2021, she discovered the unconventional superconductor cerium-rhodium-arsenic CeRh₂As₂). Superconductors normally have just one phase of resistance-free electron transport, which occurs below a certain critical temperature. However, as reported in the academic journal Science, CeRh₂As₂ is so far the only quantum material to boast two certain superconducting states.

Lossless current conduction in superconductors has remained a central focus in solid-state physics for decades and has emerged as a significant prospect for the future of power engineering. The discovery of a second superconducting phase in CeRh₂As₂, which results from an asymmetric crystal structure around the cerium atom (the rest of the crystal structure is completely symmetrical), positions this compound as a prime candidate for use in topological quantum computing.

Sudden jumps in large language models’ apparent intelligence are often a result of the way their performance is tested. Plus, a GPT-powered robot chemist designs reactions and what’s in store for AI in 2024.

NEW ORLEANS—“OK, I want to finish writing this article today. But how should it begin? Hmm, maybe just like that. Wait, does Science use ‘OK’ or ‘Okay’?” Many people say they have an inner monologue running through their heads, narrating their lives. The phenomenon, plus a wealth of research, suggests humans use language not only for communicating, but also for thinking.

Now, it seems artificial intelligence (AI) may benefit from imitating humans’ inner monologue. In a laboratory experiment, tying language to actions improved an AI program’s ability to learn complex tasks, researchers reported here last month at the Neural Information Processing Systems (NeurIPS) conference. The advance might enable AI to learn from, say, YouTube instructional videos.

“I really liked this,” says Anna Borghi, a psychologist at Sapienza University of Rome who studies cognition and language and was not involved in the experiment. “The most interesting aspect is that the presence of language grants flexibility. Even complex actions can be accomplished more easily.”