Algunos de Las mejores tecnologías que vemos en CES Parece sacado directamente de la ciencia ficción. Ayer en CES 2025, probé la tecnología AirTouch de Neural Lab, que te permite interactuar con la pantalla usando solo gestos con las manos, exactamente como son las películas. Informe de minoría y hombre de hierro una promesa. Naturalmente, muchas empresas han introducido diferentes formas de control por gestos. Microsoft Kinect Es uno de los primeros ejemplos del Apple Watch. Función de doble clic Y Visión Pro Gestos de pellizco Son sólo dos de muchas iteraciones actuales. Pero me impresionó lo bien que se entregó AirTouch y, a diferencia de la mayoría de la tecnología de gestos que existe, no requiere equipo especial (solo una cámara web estándar) y funciona con una amplia gama de dispositivos.
Neural Lab es compatible con tabletas, PC y cualquier dispositivo que ejecute al menos Android 11, Windows 10 y versiones posteriores, o Linux. La tecnología se desarrolló teniendo en cuenta la accesibilidad después de que uno de los fundadores tuviera problemas para mantenerse en contacto con sus padres en el extranjero porque navegar por el software de videoconferencia era demasiado difícil para la generación anterior. El representante de Neural Labs con el que hablé agregó que sus padres prefieren usar un iPad en lugar de una combinación de computadora, mouse y teclado porque los controles táctiles son mucho más intuitivos. Con AirTouch, pueden usar su televisor tal como usan una tableta.
Además de la accesibilidad, también existen muchas aplicaciones comerciales, como permitir a los cirujanos procesar imágenes por resonancia magnética sin tocar nada o un escenario más común como navegar a través de diapositivas en una presentación.
AirTouch rastrea los movimientos de las manos en 3D y los cambios de la mirada para reconocer la intención, lo que le permite ignorar gestos extraños. Actualmente admite nueve gestos y la personalización permite a los usuarios programar hasta 15 gestos.
Probé dos demostraciones: una pantalla 3D con una rana arbórea animada y una pantalla que muestra una página web en el navegador. En la pantalla 3D, un dedo dejó caer una piña sobre la cabeza de la rana, dos dedos dejaron caer una bellota, un pulgar hacia arriba hizo girar a la rana y el gesto tranquilo de un lobo la trajo de regreso. Me tomó 15 segundos aprender y usar los cuatro gestos y pronto estaba bañando a la pobre rana con bellotas como una ardilla de mal genio.
Fue casi tan fácil (aunque no tan divertido) controlar la pantalla que muestra el navegador web. Al mover mi mano, arrastré el cursor por la pantalla y el dial tomó el lugar del clic. Pude desplazarme por el sitio de transmisión, seleccionar algo para reproducir, pausar y reproducir nuevamente a los pocos segundos de aprender los movimientos de la mano. Hubo algunos casos en los que mis movimientos no fueron lo que esperaba, pero después de algunos intentos, comencé a dominar los controles.
AirTouch ahora está disponible como una suscripción mensual de $30 para individuos (y $300 por mes para empresas). Neural Labs dice que solo lleva cinco minutos instalar el software en cualquier dispositivo compatible.
Tengo una empresa llamada Elevation Lab. Suplemento emitido a etiqueta aérea de manzana Lo que extiende la vida útil de la batería hasta por 10 años. y Lo hace resistente al agua. el Cápsula del tiempo Básicamente es un estuche que contiene un AirTag y dos baterías AA. No es necesario abrir y manipular el AirTag; simplemente retire el panel posterior de metal y la batería antes de enchufarlo al estuche. Como puede ver, hará que su rastreador sea mucho más grande y pesado, por lo que es ideal para usar con objetos más grandes, como vehículos y maletas grandes.
Como dice Elevation Lab en su anuncio, TimeCapsule puede facilitar la administración de múltiples AirTags. Ya no tendrás que cambiar las pilas cada año sólo para asegurarte de que funcionará según lo previsto si pierdes tus cosas. Sin embargo, las baterías que uses determinarán cuánto durarán antes de que tengas que quitar la carcasa y reemplazarla.
Las baterías Energizer Ultimate Lithium recomendadas por Elevation Lab son de 1,5 voltios cada una y tienen una capacidad de más de 3000 mAh, por lo que pueden alimentar su AirTag por hasta una década. Otras baterías AA funcionarán con el estuche, pero pueden no ser tan efectivas, dependiendo de sus especificaciones. La cantidad que gastes en toda la configuración dependerá de las baterías que elijas, pero el estuche en sí te costará $20.
Amazonas Anunció el lunes la creación de un nuevo laboratorio de inteligencia artificial (IA). El nuevo laboratorio de investigación, llamado Amazon AGI SF Lab, tendrá su sede en San Francisco y se centrará en el desarrollo de agentes de IA con aplicaciones del mundo real. La tecnología con sede en Seattle contrató recientemente a varios altos ejecutivos de la startup Adept AI Labs y sembrará la nueva división con nuevos empleados. Además, el laboratorio Amazon AGI SF estará dirigido por el ex director ejecutivo y cofundador de Adept, David Loam.
Amazon crea el laboratorio Amazon AGI SF
en un Publicación de blogEl gigante tecnológico afirmó que la nueva unidad de investigación se centrará en desarrollar capacidades centrales para agentes de inteligencia artificial que puedan actuar en los mundos digital y físico. En particular, los agentes de IA pueden entenderse como chatbots de IA más pequeños y eficientes que también pueden realizar acciones mediante la integración con hardware y software especializados.
Amazon enfatizó que la responsabilidad del equipo incluirá la construcción de “inteligencia artificial práctica” que pueda realizar tareas tanto para la empresa como para sus clientes. Como sugiere el nombre de la división, el equipo también trabajará en estrecha colaboración con el equipo de Inteligencia General Artificial (AGI) de Amazon, que recientemente presentó la familia Nova de modelos fundamentales.
“Nuestro enfoque inicial está en varias apuestas de investigación clave que permitirán a los agentes de IA realizar acciones en el mundo real, aprender de los comentarios humanos, autocorregirse e inferir nuestros objetivos”, afirmó la compañía en la publicación del blog. Amazon también destacó que el equipo combinará grandes modelos de lenguaje (LLM) y aprendizaje por refuerzo (RL) para resolver inferencias y planificación, modelos de mundo aprendido y generalizar agentes a entornos físicos.
Amazon también anunció que está buscando contratar “unas pocas docenas” de personas para Amazon AGI SF Lab. La empresa busca expertos en IA que hayan entrenado modelos de última generación (SOTA), así como candidatos de otros campos como física, matemáticas, finanzas cuantitativas y otros. La publicación del blog también agregó que el nivel de experiencia no es un criterio para la contratación.
La semana pasada, Amazon Web Services (AWS) pie La familia de modelos de inteligencia artificial (IA) de Nova en su conferencia re:Invent, con tres modelos basados en texto, un modelo de generación de imágenes y un modelo de generación de video. Todos estos modelos están disponibles para los clientes empresariales de la empresa y pueden aprovecharse desde la plataforma Amazon Bedrock.
Apple desarrolló su experiencia de salud auditiva líder en la industria para AirPods Pro 2 en su laboratorio de audio de última generación en Cupertino. Esta innovación, disponible en una actualización de software gratuita, llega en un momento crítico. Después de todo, la Organización Mundial de la Salud informa que aproximadamente 1.500 millones de personas en todo el mundo sufren pérdida de audición.
Como dijo un ingeniero de Apple, con las nuevas funciones de salud auditiva, los AirPods Pro 2 se convierten en “verdaderamente la interfaz del oído”.
Inside Apple Audio Lab: Liderazgo en tecnología accesible para la salud auditiva
el Nuevas características Desarrollado en Apple Audio Lab, disponible como archivo Actualización de software gratuita en iOS 18.1 y iPadOS 18.1Incluye tres componentes principales. Se trata de protección auditiva para reducir la exposición a ruidos ambientales fuertes, pruebas de audición en el hogar similares a las que se realizarían en el consultorio de un médico y audífonos de grado clínico diseñados para pérdidas auditivas de leves a moderadas.
“La audición de cada persona es diferente, por eso creamos una experiencia de salud auditiva integral e innovadora que es fácil de usar y adaptable a una amplia gama de necesidades”, explicó el Dr. Sombul Desai, vicepresidente senior de Salud de Apple. “Con Audición, queríamos crear algo tan intuitivo que pareciera una extensión de los sentidos”.
Cámara anecoica de onda larga y Fantasia Lab dentro del laboratorio de audio de Apple
Los usuarios de AirPods Pro 2 con conectores Lightning o USB-C pueden descargar y utilizar las funciones de Salud Auditiva. Imagen: manzana
El proceso de desarrollo utilizó varias instalaciones especializadas dentro del Audio Lab de Apple. La cámara anecoica de onda larga, construida sobre una base separada montada sobre resortes, permite mediciones acústicas precisas sin interferencias externas. Este espacio único cuenta con un arco de altavoz y micrófono diseñado específicamente para medir cómo el sonido interactúa con el cuerpo humano.
Otra instalación importante, el Fantasia Lab, utiliza 50 altavoces dispuestos esféricamente para simular cientos de entornos acústicos del mundo real, desde centros comerciales hasta calles concurridas. Esta configuración era necesaria para probar y perfeccionar la ventaja auditiva, ya que los participantes del estudio de varios grupos demográficos participaron en pruebas de habla en ruido.
En el laboratorio, los simuladores de cabeza y torso miden los AirPods y validan la prueba de audición y las funciones del audífono. Imagen: manzana
El laboratorio también incluye tres cabinas de audiometría de nivel clínico, donde los ingenieros colaboraron con audiólogos para realizar miles de pruebas de audición antes de pasar a estudios de validación clínica. El equipo de diseño se centró en crear una experiencia de usuario intuitiva, especialmente para la función de prueba de audición, haciéndola más accesible que las evaluaciones clínicas de audición tradicionales.
“Dentro de nuestras funciones de salud, nos centramos en la claridad y en encontrar a los usuarios donde estén”, dijo Heather Daniel del Estudio de Diseño de Apple. “Nos dimos cuenta de que para muchas personas esta podría ser la primera vez que se sometieran a una prueba de audición, por lo que tuvimos que hacerlo lo más fluido posible”.
Los equipos de Apple colaboran para crear una 'interfaz auditiva'
Para ajustar y validar la ventaja auditiva, los participantes con una amplia gama de niveles auditivos completaron una prueba de habla en ruido en un entorno controlado. Imagen: manzana
El proyecto requirió una amplia colaboración entre varios equipos de Apple, incluida la ingeniería de software y hardware, diseño, salud, accesibilidad, operaciones clínicas e ingeniería de factores organizativos y humanos.
“Tus oídos son parlantes naturales, cada uno con una forma única y, a menudo, ligeramente asimétrica. Cuando el sonido llega primero a un oído antes que al otro, se crea una diferencia temporal en la forma en que lo percibimos”, dice Koba Mazur, ingeniero senior de salud auditiva de Apple. “Es importante para nosotros entender esto para poder crear experiencias que representen con precisión los sonidos de su entorno”.
Para simplificar las pruebas de audición en AirPods Pro 2, el equipo de diseño llevó a cabo una investigación exhaustiva para determinar qué siente una persona cuando se somete a una prueba de audición tradicional para identificar áreas que pueden simplificarse para cualquier usuario que realice la prueba en casa. Imagen: manzana
Las nuevas funciones de salud auditiva representan el compromiso continuo de Apple con la innovación en tecnología sanitaria. La incorporación de estas capacidades a los AirPods Pro 2 demuestra cómo los dispositivos cotidianos pueden transformarse en poderosas herramientas de salud. Esto podría ayudar a que la tecnología avanzada de salud auditiva sea accesible para millones de usuarios.
Como concluye Mazur: “El hecho de que las personas puedan caminar usando sus AirPods, proteger su audición en conciertos y obtener información sobre la salud auditiva utilizando estas funciones con el tiempo, los AirPods hacen lo que cada persona quiere o necesita hacer”. La interfaz es para el oído”.
Aparentemente era un hecho que las películas de MCU se harían con CGI tan caro (y apresurado) como fuera posible. Después de todo, el Universo Marvel es un lugar de fantasía lleno de robots, psíquicos, magos y piedras mágicas superpoderosos que controlan la realidad. Es lógico que se necesitaran muchos efectos visuales para lograr esto en acción real. Sin embargo, muchos realizadores de MCU pueden haber abusado de los efectos visuales, a veces utilizando fondos CGI o personajes animados cuando el maquillaje o los decorados prácticos podrían verse mejor. Reid, que quería que su película pareciera un poco más tangible que sus pares, construyó un escenario enorme, permitiendo a los actores explorar un espacio físico de manera orgánica:
“Han creado este nuevo laboratorio en el que Hope y Hank están trabajando en este increíble túnel cuántico. Túnel cuántico, debo agradecer a Shepard Frankel, nuestro diseñador de producción. Este es el conjunto de física más grande jamás creado que fue diseñado específicamente para una película de Marvel. Es un poco contradictorio que Ant-Man and The Wasp tenga el escenario físico más grande.
Reid fue perspicaz cuando notó la ironía.
Shepard Frankel comenzó a trabajar como diseñador de producción en comerciales a principios de la década de 2000 y consiguió su primer trabajo cinematográfico con “Step Up” en 2006. Posteriormente trabajó en varias comedias de Hollywood de alto perfil, estableciéndose como una presencia talentosa y confiable en la industria. industria del entretenimiento. Su primer proyecto con Marvel fue el cortometraje de 2012 “Item 47”, ambientado después de “Los Vengadores”. También diseñó los cortos de Marvel “Agent Carter” y “All Hail the King” antes de ser contratado por Reed para la primera película de “Ant-Man”. Otras obras de Frankl incluyen Los elaborados diseños de “Uncharted”, Además del próximo “Harold and the Purple Crayon”.
Karolina Makovskytė is head of business development at the firm Caszyme, which specializes in CRISPR-based molecular tools.Credit: Caszyme, LLC archive
Voices from Lithuania
In May, Lithuania marks 20 years of European Union membership. The Baltic country is keen to develop its global presence in the life sciences and biotechnology sectors by retaining home-grown talent, persuading scientists working abroad to return to the country, and attracting researchers from other nations. Naturespoke to three researchers who have chosen to develop their careers in Lithuania. Here, Nature speaks to Karolina Makovskytė, head of business development at Caszyme, a biotechnology firm that specializes in CRISPR-based molecular tools.
Before the COVID-19 pandemic, when I was an undergraduate molecular-biology student at Vilnius University in Lithuania, I loved laboratory work. But when labs closed during the pandemic, I lost my motivation and it never returned.
In 2021, during the fourth and final year of my degree, a recruiter contacted me about a part-time junior product-manager post at the Vilnius site of biotechnology firm Thermo Fisher Scientific. The company was manufacturing reagents for messenger RNA vaccines, and it had built a whole new facility and was producing them on a large scale.
Thermo arrived in the country in 2010, when it acquired Lithuanian biotech company Fermentas for US$260 million. Its Vilnius site manufactures products for the life-science market, specifically molecular, protein and cellular-biology products. It also has a research and development centre.
It’s common for science students to work during their studies and for me it provided a chance to see what working in the biotech field was like, and to find out what options I had with a life-sciences degree, other than working in the laboratory.
I was working on a portfolio of nucleic-acid therapeutics, helping senior product managers with requests from companies for different nucleotides and enzymes, mainly for mRNA vaccine development and production. I carried on working at Thermo, during my master’s degree in molecular biotechnology, also at Vilnius University. Even on the administrative and business side, many people had a master’s or PhD.
As part of my master’s, I attended lectures on the genome-editing tool CRISPR, led by Giedrius Gasiūnas, co-founder and chief scientific officer at biotech company Caszyme in Vilnius.
Virginijus Šikšnys, another co-founder and chair of the management board, is a department head at Vilnius University’s Institute of Biotechnology. This demonstrates the blended nature of academia and the world of biotech start-up companies in Lithuania.
I applied for a business-development manager role there and started work in March 2022, while doing my master’s degree, which I completed last year. My job involves relationship building and representing the company abroad.
The company discovers and develops CRISPR-based molecular tools for diverse applications. It has 17 scientists and is growing all the time. In January, we moved premises to a larger building opposite the Vilnius site of the firm Northway Biotech, and close to Vilnius University Hospital.
Currently, Caszyme focuses on three elements. The first is research: finding new CRISPR–Cas nucleases for companies that are looking for specific characteristics. Second is licences: the company has a platform of nucleases that it has already identified, and it offers licences for further development. And third is analysis: Caszyme helps companies with expression, purification and biochemical analysis of Cas nucleases.
Currently, the firm has a handful of active collaborations in the therapeutics and diagnostics fields, including in infectious diseases and agriculture.
A friendly and close community
I think there are many aspects of academia and industry that set Lithuania apart from other countries. First, we have great universities that prepare scientists very well for the future. Scientists leave with not only theoretical knowledge, but also research experience working in a lab.
Karolina Makovskytė moved into industry after discovering the lab was not for her.Credit: Caszyme, LLC archive
Second is that the ecosystem is not a big one. It’s very easy to get to know most people and that’s an advantage because it makes collaborations easier.
Third, industry scientists have close contact with government institutions that ask us what businesses want and how they can help. I don’t want to say it’s easy to build a biotechnology company, but Lithuania has all the right tools to do high-quality research.
My proudest achievement is probably my personal growth. I can really see a difference between now and when I started working — I’ve gained knowledge and experience, and matured as a person. Last June, aged 25, I was promoted to head of business development.
I work remotely in Kaunas, the second-largest city in Lithuania, which has a beautiful old town and is known for its art deco architecture. We have three international airports and Vilnius is only an hour and a half by road, so it’s easy to get to the office when I need to. Some of my time is spent abroad — I’ve just come back from a technology summit in San Francisco, California, and now I’m in Rome.
I don’t want to work abroad permanently. Lithuania has wonderful countryside and is full of culture, too. Every year, in March we have an international film festival, which I have worked for as a volunteer in the past.
If you find you don’t like working in a lab, like me, my advice would be to find a job outside the lab and see if it fits. If it doesn’t, try something else and talk to people who have different positions in life sciences. Explore the opportunities.
I talk to lots of people at events and many of them have PhDs or have worked as postdocs but have decided that they don’t want to be scientists. There’s nothing shameful in that. People who work in business development or operations and who have a scientific degree are essential, too.
My advice for young scientists who are considering a research position in Lithuania is to just do it. We have a friendly and close research community so it’s very easy to collaborate. Also, scientists here have a wide network all over the world. It’s very easy to join projects that Lithuanian scientists are doing with global institutes and companies. We Lithuanians talk to each other a lot!
Jacqui Thornton’s travel and accommodation were provided by Go Vilnius, a development agency.
This interview has been edited for length and clarity.
This article is part of Nature Spotlight: Lithuania, an editorially independent supplement. Advertisers have no influence over the content.
The first person to publicly release the genome sequence of the virus that causes COVID-19 — virologist Zhang Yongzhen — appears to have resolved a public dispute with the Shanghai Public Health Clinical Center (SPHCC), Fudan University, that erupted last week.
Zhang had been photographed camping outside his laboratory since Sunday, after the SPHCC closed the lab.
The SPHCC did not respond to Nature’s request to comment but has released public statements saying the laboratory was one of several that needed renovations — and that it provided Zhang’s team with an alternative space.
According to social media posts on Zhang’s personal Weibo account, the institute gave the research team two days to leave but the SPHCC did not initially specify where they should relocate. Later, Zhang said officials told his team to move to a lab that did not have the necessary biosafety conditions to store their samples, which contain unknown pathogens. Zhang’s lab is a biosafety level 3 laboratory.
Zhang said he had been sleeping outside his laboratory since Sunday night, even in the rain. The social media posts include photos of him lying under blankets.
Zhang told Nature on Monday that his situation was “terrible”.
“You don’t know what I have experienced,” he said, but declined to comment further.
According to the social media account of Chen Yanmei, a virologist at the SPHCC, and a member of Zhang’s team, their students’ incomplete experiments were now “impossible to save”. Chen said she was also camping out, inside the lab. Chen also declined to be interviewed by Nature.
But by late Tuesday night, Zhang said in a post that a tentative agreement had been reached with the SPHCC to resume normal research activity in the laboratory. The post states that Zhang will work with the centre to relocate the laboratory and restart research.
Virus sequence
In 2020, Zhang was the first scientist — together with Edward Holmes, a virologist at the University of Sydney in Australia — to share the genome of SARS-CoV-2 with the world on the website virological.org. That act is widely credited as a key factor enabling the swift development of vaccines against COVID-19.
Since 2020, Zhang has received international acclaim. That year, Zhang featured in Nature’s 10 — an annual list of people behind key developments in science, and in 2022 he was awarded the Mohammad bin Rashid Al Maktoum Knowledge Award, which carries a US$1 million bursary.
Nature’s 10: ten people who helped shape science in 2020
But Holmes, a long-time collaborator of Zhang’s, says that since 2020, Zhang’s research output has dwindled. Before 2020, Zhang had an extensive research network and would collect samples of animals and people to study viral evolution, says Holmes. But since 2020, Zhang’s work has largely involved analysing previously collected samples and Holmes’s own collaboration with Zhang is less prolific. “He drove that collaboration but there is nothing to collaborate on now; he hasn’t been able to get any data,” says Holmes. “All I can do is offer support from afar.”
According to the Dimensions database, Zhang co-authored 5 research articles in 2018, 9 in 2019 and 18 in 2020. But that growing publication rate dropped to 3 in 2021 and 4 in 2022. Zhang co-authored 6 articles in 2023, none of which contained original data.
Holmes believes the lab closure is part of an effort to sideline Zhang for unauthorized sharing of data. “It is heartbreaking to watch,” he says. “It is unfathomable to me to have a scientist of that calibre sleeping outside his lab.”
International star
But Yanzhong Huang, a specialist in Chinese health policy at the Council on Foreign Relations in New York City, says the true nature of the dispute and protest are “shrouded in mystery”.
Fan Xiaohong, a physician who heads the SPHCC told reporters for Chinese newspaper Southern Weekly that Zhang’s contract had expired, but he had refused to leave. A post on Zhang’s social media said that although his own contract with SPHCC had formally ended in 2023, members of his team had renewed their contracts with the hospital, and the laboratory is still owed funding.
Even without clarity on the details of the dispute, Dali Yang, a researcher who studies politics in China at the University of Chicago, says it’s concerning that Zhang is sleeping outside his lab. “Many people who know about him are aghast because he is truly an international star,” says Yang.
Apple has poached dozens of artificial intelligence experts from Google and created a “secretive European laboratory” in Zurich to house a new team of staff tasked with building new AI models and products, according to a paywalled Financial Times report.
Based on an analysis of LinkedIn profiles conducted by FT, Apple has recruited at least 36 specialists from Google since 2018, when it poached John Giannandrea to be its top AI executive.
Apple’s main AI team works out of California and Seattle, but the company has recently expanded offices dedicated to AI work in Zurich, Switzerland. Apple’s acquisition of local AI startups FaceShift (VR) and Fashwell (image recognition) is believed to have influenced its decision to build a secretive research lab known as “Vision Lab” in the city.
According to the report, employees based in the lab have been involved in Apple’s research into the underlying technology that powers OpenAI’s ChatGPT chatbot and similar products based on large language models (LLMs). The focus has been on designing more advanced AI models that incorporate text and visual inputs to produce responses to queries.
The report suggests that Apple’s recent work on LLMs is a natural outgrowth of the company’s work on Siri over the last decade:
The company has long been aware of the potential of “neural networks” — a form of AI inspired by the way neurons interact in the human brain and a technology that underpins breakthrough products such as ChatGPT.
Chuck Wooters, an expert in conversational AI and LLMs who joined Apple in December 2013 and worked on Siri for almost two years, said: “During the time that I was there, one of the pushes that was happening in the Siri group was to move to a neural architecture for speech recognition. Even back then, before large language models took off, they were huge advocates of neural networks.”
Currently, Apple’s leading AI group includes notable ex-Google personnel such as Giannandrea, former head of Google Brain, which is now part of DeepMind. Samy Bengio, now senior director of AI and ML research at Apple, was also previously a leading AI scientist at Google. The same goes for Ruoming Pang, who directs Apple’s “Foundation Models” team focusing on large language models. Pang previously headed AI speech recognition research at Google.
In 2016, Apple acquired Perceptual Machines, a company that worked on generative AI-powered image, detection, founded by Ruslan Salakhutdinov from Carnegie Mellon University. Salakhutdinov is said to be a key figure in the history of neural networks, and studied at the University of Toronto under the “godfather” of the technology, Geoffrey Hinton, who left Google last year citing concerns about the dangers of generative AI.
Salakhutdinov told FT that one reason for Apple’s slow AI rollout was the tendency of language models to provide incorrect or problematic answers: “I think they are just being a little bit more cautious because they can’t release something they can’t fully control,” he said.
A first look at the AI features that Apple has planned should come in just over a month, with iOS 18 set to debut at the Worldwide Developers Conference that kicks off on June 10.
Superconductivity has been demonstrated at extremely low temperatures, but it remains elusive at room temperatures.Credit: Brookhaven National Laboratory/SPL
Research misconduct is hugely detrimental to science and to society. Defined as “fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results” by the US Office of Research Integrity, it violates trust in science and can do great harm to the wider public, scientific institutions and especially co-authors and students who had no part in the wrongdoing. In cases involving public funds, it squanders resources that could have been allocated to other research and it can erode lawmakers’ support for science.
Does the scientific community, as a whole, have appropriate processes for reporting, investigating and communicating about instances of potential misconduct? This question is not new. At Nature, we’re asking it again, after two separate studies that we published were subsequently retracted.
Exclusive: official investigation reveals how superconductivity physicist faked blockbuster results
The studies1,2 were originally published in October 2020 and March 2023. The first was retracted in September 2022 and the second in November 2023. The corresponding author on both papers was Ranga Dias, a physicist studying superconductivity at the University of Rochester in New York, and a recipient of grants from the US National Science Foundation (NSF).
The papers by Dias and his co-authors claimed to report room-temperature superconductivity under extremely high pressures, each in different materials. Room-temperature superconducting materials are highly sought after. They could, for example, transform the efficiency of electricity transmission, from the smallest to the largest application. But high-pressure experiments are difficult and replicating them is complex.
Nature initiated an investigative process that resulted in the 2020 paper being retracted after members of the community told the journal they were troubled by aspects of the data being reported. Nature also initiated an investigation into the 2023 paper. However, this article was retracted at the request of most of Dias’s co-authors while the investigation was still ongoing.
Many details about this case came to light thanks to continued questions from the research community, including during post-publication peer review. Much credit must also go to the persistence of science journalists, including members of Nature’s news team (which is editorially independent of Nature’s journal team) and those from other publications.
What can journal editors, funding organizations and institutions that employ researchers learn from such cases? We have the same goal: producing and reporting rigorous research of the highest possible standard. And we need to learn some collective lessons — including on the exchange of information.
The University of Rochester conducted three inquiries, which are a preliminary step to making a decision about whether to perform a formal investigation into scientific misconduct. The inquiries were completed between January and October 2022. Each concluded that such an investigation was not warranted.
Superconductivity scandal: the inside story of deception in a rising star’s physics lab
Earlier this month, Nature’s news team uncovered a 124-page report on a subsequent confidential investigation, performed at the NSF’s request. In it, a team of reviewers concluded after a ten-month assessment of evidence that it was more likely than not that Dias had committed data fabrication, falsification and plagiarism. The report is dated 8 February 2024, and the determination is regarding the two Nature papers, a 2021 study3 published in Physical Review Letters and a 2022 study4 in Chemical Communications — both of which were also retracted. However, the investigation has not yet officially been made public.
Some researchers have asked why Nature published Dias’s second paper in March 2023, when questions were being asked about the first one. Others have asked why the retraction notices didn’t spell out that there has been misconduct.
It’s important to emphasize that it’s Nature’s editorial policy to consider each submission in its own right. Second, peer review is not designed to identify potential misconduct. The role of a journal in such situations is to correct the scientific literature; it is for the institutions involved to determine whether there has been misconduct, and to do so only after the completion of due process, which involves a systematic evaluation of primary evidence, such as unmodified experimental data.
Access to raw data is fundamental to resolving cases of potential misconduct. It is also something we constantly think about in relation to publishing. Indeed, for certain kinds of data, Nature requires authors to deposit them in external databases before publication. But there must be more the research community — including funders and institutions — can all do to incentivize data sharing.
Another question is whether the matter could have been dealt with more quickly. Nature’s editors have been asking the same question: specifically, could there have been more, or better, communication between journals and institutions once evidence of potential misconduct came to light?
Publish, and be damned…
Last month, the Committee on Publication Ethics (COPE), a non-profit organization that represents editors, publishers and research institutions, updated its guidelines on how publishers and universities could communicate better. The guidelines are full of important advice, including that institutions, not publishers, should perform integrity or misconduct investigations. Investigators require access to primary evidence. As employers and grant-givers, institutions are the appropriate bodies to mandate access to unmodified experimental data, correspondence, notebooks and computers and to interview relevant staff members — all essential parts of an investigation.
But often, journals need to start a process that could lead to retracting a study in the absence of an institutional investigation — or while an investigation, or inquiry, is ongoing5. Are cases such as this an opportunity for journals and institutions to discuss establishing channels through which to exchange information, in the interest of expedited outcomes — as part of due process? Nature’s editors would be willing to play a part in such discussions.
Retractions are part of publishing research, and all journals must be committed to retracting papers after due process is completed. Although a paper can be retracted for many reasons, when the cause is potential misconduct, institutions must conduct thorough investigations.
This case is not yet closed. Both the university and the funder need to formally announce the investigation’s results, and what action they intend to take. They should not delay any more than is necessary. When there is credible evidence of potential scientific misconduct, investigations should not be postponed. There is strength in collaborating to solve a problem, and nothing to be ashamed of in preserving the integrity of the scientific record.
A photographic archive has been discovered in Lyon, France, that adds precious detail to what we know about the founding of the world’s first police crime laboratory in 1910 and its creator, Edmond Locard, a pioneer of forensic science.
The huge collection, which comprises more than 20,000 glass photographic plates that document the laboratory’s pioneering scientific methods, crime scenes and Locard’s personal correspondence, is thrilling historians at a time when many consider that forensic science has lost its way. “There is a movement to look back to the past for guidance as to how to renew the science of policing,” says Amos Frappa, a historian affiliated with the Sociological Research Centre on Law and Criminal Institutions in Paris, who is overseeing the analysis of the images.
She was convicted of killing her four children. Could a gene mutation set her free?
In the late nineteenth and early twentieth centuries, many people in Europe and beyond were thinking about how criminals might be accurately identified by using techniques such as fingerprint, blood and skeletal analysis. Locard was the first person to create the semblance of forensic science. He established the first scientific lab that came under the aegis of the police, and that was dedicated to studying ‘traces’ of criminal activity collected from crime scenes.
Garage find
The collection of photographic plates almost didn’t survive. It languished for decades in a garage belonging to the National Forensic Police Department in Ecully, a Lyon suburb. In 2005, the glass plates were rescued from the garage and stored in Lyon’s municipal archives. But at the time, the Lyon archives lacked the resources to treat the collection properly, says director Louis Faivre d’Arcier. It wasn’t until 2017 that an inspection revealed that the plates’ gelatine layer containing the image information was, in many cases, infected with mould. After a sorting and decontamination project in 2022, conservators saved around two-thirds of the plates.
Left: A tattooed woman named Marie-Clémentine in 1934; Edmond Locard’s team used tattoos as a way of identifying potential criminals. Right: Handwriting analysis as a means of identification was investigated but later spurned by Locard, who deemed it unreliable.Credit: Archives municipales de Lyon
The mammoth task of digitizing the contents of the fragile plates, which are mostly unindexed and disordered, became possible only when a local publisher and historian of funerary practices, Nicolas Delestre, offered to finance it. In collaboration with the municipal archives, his team developed a photographic protocol to capture as much information from the plates as possible. The digitization will be completed this spring, to coincide with the publication of Frappa’s French-language biography of Locard. The slow rebuilding of the indexes continues.
Locard, who worked in the early to mid-twentieth century, is famous for his maxim, which is usually formulated in English as “Every contact leaves a trace.” Trained as a forensic pathologist, he turned to the study of trace evidence after a French political scandal called the Dreyfus affair, in which a Jewish army officer called Alfred Dreyfus was falsely accused of espionage. During the affair, Locard’s mentor Alphonse Bertillon, who had invented a method of identifying people through bodily measurements, was called on as a handwriting specialist, despite having no expertise in the field. He wrongly identified Dreyfus as the author of an incriminating note.
Forensic science: The soil sleuth
Locard, seeing other countries adopt fingerprint identification, embraced that method instead. In 1910, he set up his laboratory in the attic of Lyon’s main courthouse, and gradually expanded his scientific analyses to include traces such as blood, hair, dust and pollen.
Sherlock Holmes connection
This much was known from published sources, but the photographic archive offers details about the social and intellectual milieu that produced Locard, onthe scientific networks in which he was embedded, and on how his thinking evolved as he experimented and made errors. His exchanges with contemporaries in countries including Germany, Switzerland, Italy and the United States shaped his approach, which might be why he did not consider himself a founder of a new field. But Locard’s ideas — his scientific methods and his insistence on meticulously studying crime scenes — fell on fertile ground in Lyon’s police chiefs and judges, who, unlike their Parisian counterparts, accepted the evidence that such approaches generated. “Lyon was a receptacle,” says Frappa.
Edmond Locard using a photographic bench in the 1920s.Credit: Archives municipales de Lyon
The new collection reveals Locard’s team at work. It captures their equipment and experiments, and the forensic traces they analysed. The close-knit group socialized together, received international visitors and investigated myriad means by which people could be identified. One way was to look at people’s tattoos, and the collection contains a large set of tattoo images. Locard took inspiration from many sources, including the Lyon-based Lumière brothers, who were pioneers of cinematography, and the creator of the fictional detective Sherlock Holmes, Arthur Conan Doyle, with whom he corresponded. In time, Locard discarded some techniques — notably, handwriting analysis — deeming them unreliable.
Since 2009, when a report from the US National Research Council found that many modern forensic techniques were inadequately grounded in science, the discipline has struggled to reorient itself. “By the late 20th century, it’s fair to say that forensic science had become an adjunct of law enforcement without allegiance to science,” says Simon Cole, who studies criminology, law and society at the University of California, Irvine, and directs the US National Registry of Exonerations. Cole has written about the problems with fingerprint identification, and last year reported on the fallibility of microscopic hair comparison. These techniques are routinely used to investigate crimes in the United States and elsewhere, and the evidence they generate is admissible in court.
Modern troubles
The 2009 report suggested that improving forensic science would require larger labs in which diverse specialists were insulated from each other and from the police to prevent bias. The trouble with that view, says Olivier Ribaux, director of the School of Criminal Sciences at the University of Lausanne in Switzerland, is that, when considering the potentially infinite number of traces that a crime scene can generate, some subjective selection by humans is inevitable. To ensure that this selection is as informative and as unbiased as possible, the forensic scientist must understand a trace in its context — as Locard’s maxim in French originally implied. “The problem with the big labs is that they have severed the connection with the crime scene,” Ribaux says.
He favours an alternative model in which smaller labs employ generalists, who can oversee specialists in certain fields, such as ballistics and DNA, but can also offer a more holistic view of a case. These generalists would work closely with the police — a return to Locard’s approach, in other words. But the two aren’t mutually exclusive, Ribaux says. They are just snapshots of the ongoing debate about how the field should reinvent itself.
That debate will surely be fuelled by the emerging portrait of Locard, sometimes dubbed the French Sherlock Holmes, whom Frappa describes as “a man so visionary he predicted, correctly, that he would be forgotten”.