MITs

Welcome, folks, to Week in Review (WiR), TechCrunch’s regular newsletter that recaps the week in tech that was. Hope the holidays were restful for those who observed them. We at TC, for our parts, are gearing up for an eventful next week at CES in Las Vegas — while keeping an eye on the news cycle, as ever. In this edition of WiR, we spotlight Brian’s CES 2024 preview, 23andMe blaming victims for its data breach, GitHub making Copilot Chat generally available and Frontdesk laying off its entire staff. Also on the…

MIT researchers propose a theoretical model explaining how cells maintain their identity over generations. The model suggests that a cell’s 3D genome structure guides the restoration of epigenetic marks lost during cell division. This mechanism enables cells to remember their specific type, with implications for understanding diseases and aging processes. Credit: SciTechDaily.comMIT study suggests 3D folding of the genome is key to cells’ ability to store and pass on “memories” of which genes they should express.Every…

MIT researchers have created a wearable ultrasound patch that can non-invasively image internal organs, initially focusing on bladder health. This device, which does not require an ultrasound operator or gel, has the potential to revolutionize the monitoring of various organ functions and disease detection.The wearable device, designed to monitor bladder and kidney health, could be adapted for earlier diagnosis of cancers deep within the body.MIT researchers have designed a wearable ultrasound monitor, in the form of a…

The MIT Game Lab combines research and education to critically explore the gaming industry and its societal impact. It encourages students to think beyond games as entertainment, focusing on their cultural significance and potential for addressing real-world issues.This unique lab uses games as a way for students to play, explore, and learn to think critically about the role of games in society.Students fill the glass-walled room and spill out into the common area. They gather around tables and desks cluttered with board…

MIT’s StableRep system uses synthetic images from text-to-image models for machine learning, surpassing traditional real-image methods. It offers a deeper understanding of concepts and cost-effective training but faces challenges like potential biases and the need for initial real data training.MIT CSAIL researchers innovate with synthetic imagery to train AI, paving the way for more efficient and bias-reduced machine learning.Data is the new soil, and in this fertile new ground, MIT researchers are planting more than…

Researchers at MIT have designed a wearable ultrasound patch that can image organs as well as a conventional ultrasound does, without the need for cold gel or an operator. While they used it to measure bladder fullness, the device could also be adapted to image other internal organs, providing a new way of monitoring for disease.Ultrasound is widely used in clinical practice. The imaging technique is painless, non-invasive, does not use ionizing radiation, and provides images in real-time. However, currently, ultrasounds…

Dall-E and Stable Diffusion were only the beginning. As generative AI systems proliferate and companies work to differentiate their offerings from those of their competitors, chatbots across the internet are gaining the power to edit images — as well as create them — with the likes of Shutterstock and Adobe leading the way. But with those new AI-empowered capabilities come familiar pitfalls, like the unauthorized manipulation of, or outright theft of, existing online artwork and images. Watermarking techniques can help…

MIT-Watson AI Lab’s new AI system drastically streamlines drug and material discovery by accurately predicting molecular properties with minimal data. The system leverages a “molecular grammar” learned via reinforcement learning to generate new molecules efficiently. This method has shown remarkable efficacy even with datasets of less than 100 samples.This AI system only needs a small amount of data to predict molecular properties, which could speed up drug discovery and material development.Discovering new materials and…

The FrameDiff system was tested on the task of building single proteins, and the researchers found that it can create big proteins with up to 500 parts. Unlike previous methods, it doesn’t need to rely on a preexisting map of the protein structure. Credit: Alex Shipps/MIT CSAIL via Midjourney<span class="glossaryLink" aria-describedby="tt" data-cmtooltip="<div class=glossaryItemTitle>MIT</div><div class=glossaryItemBody>MIT is an acronym for the Massachusetts Institute of…

MIT-Takeda Program researchers have developed a physics and machine learning technique to enhance the manufacturing process of pharmaceutical pills and powders. Their method, called PEACE, involves using a laser and machine learning to measure particle size distribution, increasing efficiency, reducing failed batches, and making the process more sustainable and cost-effective.A collaborative research team from the <span class="glossaryLink" aria-describedby="tt" data-cmtooltip="<div…