Subscribe to Updates
Subscribe to our newsletter and never miss our latest news
Subscribe my Newsletter for New Posts & tips Let's stay updated!
Author: TechBrunch
The U.S. Department of Commerce today released a report supporting “open-weighted” generative AI models like Meta's Llama 3.1, but recommending that the government develop “new capabilities” to monitor these models for potential risks. The report, written by the Commerce Department's National Telecommunications and Information Administration (NTIA), argues that open-weighted models would broaden the scope of generative AI available to small businesses, researchers, nonprofits, and individual developers. For these reasons, the report suggests, governments should not restrict access to open models, at least not before studying whether such restrictions would have a negative impact on the market. This sentiment echoes recent…
Shared micromobility giant Lime is piloting two new vehicles designed for women and seniors who prefer a lower step-through frame, smaller wheels and an upgrade from pedal assist. Lime has tried several new form factors over the years to win over customers and adapt to changing tastes and trends. The efforts have paid off in some important ways. The company has been able to turn a profit at times in an industry where most other companies, such as Tier, Bird, and Spin, have struggled to survive. In 2023, Lime earned more than $90 million in adjusted EBITDA, which the company…
Apple has published a technical paper detailing the models it developed to power Apple Intelligence, a set of generative AI features coming to iOS, macOS, and iPadOS in the coming months. In the paper, Apple refutes accusations that it took an ethically questionable approach to training some of its models, reiterating that it does not use private user data and that it uses a combination of public and licensed data for Apple Intelligence. “[The] “The pre-training datasets consist of data licensed from publishers, curated public or open-source datasets, and public information crawled by our web crawler, Applebot,” Apple wrote in…
A chat between Nvidia CEO Jensen Huang and Meta CEO Mark Zuckerberg at the SIGGRAPH 2024 conference in Colorado on Monday took an unexpected turn. The conversation started out as standard — Huang spoke positively about the power of Nvidia GPUs and Zuckerberg talked about the future of AI chatbots — but ended with Huang and Zuckerberg exchanging custom jackets, a conversation reminiscent of a meme that went viral earlier this year. “It's black and leather and shearling,” Zuckerberg said of the coat he brought for Huang. “It's got vibe. Buy this guy a chain.” Image credit: Nvidia Huang gifted…
Meta had a big win last year with “Segment Anything,” a machine learning model that could quickly and reliably identify and outline almost anything in an image. The sequel, which CEO Mark Zuckerberg debuted onstage at SIGGRAPH on Monday, brings the model into the video realm, showing how rapidly the field is evolving. Segmentation is the technical term for a visual model looking at an image and picking out parts: “This is a dog, and this is the tree behind the dog,” and hopefully not “This is the tree growing out of the dog.” This has been done for decades,…
Henrik Fisker once envisioned a burgeoning electric-vehicle empire led by the Ocean SUV at his eponymous startup, but cracks in that vision started to show as soon as the Ocean hit the road in 2023. Fisker repeatedly cut production goals, missed sales targets, and laid off employees. Additionally, its Ocean SUVs were hit by software and mechanical problems that rendered them inoperable for some. Problematic brakes, sudden loss of power, and doors that wouldn't open added to a list of issues that led to multiple safety investigations and ultimately a production halt to raise new capital. These and other factors…
So-called “unlearning” techniques are used to make generative AI models forget certain undesirable information taken from the training data, such as sensitive personal data or copyrighted material. But current unlearning techniques are a double-edged sword: they can significantly degrade the ability of models like OpenAI's GPT-4o and Meta's Llama 3.1 405B to answer basic questions. That's according to a new study co-authored by researchers from the University of Washington (UW), Princeton University, University of Chicago, University of Southern California, and Google, which finds that today's most popular unlearning techniques tend to degrade models, often to the point where they become…
Uber is rolling out simultaneous rides in India, a feature that lets users book up to three rides for anyone in their contacts, according to information obtained exclusively by TechCrunch and confirmed by the company. The co-ride feature is the latest example of how Uber is developing its products to reach more customers, including those who don't have the app or a smartphone. In India, Uber also allows co-ride users to pay the driver directly with cash or via the app. Uber quietly launched its simultaneous ride-hailing service in markets around the world, including the U.S., last year. An Uber…
US airports are now rolling out facial recognition technology to scan travelers' faces before they board – and Americans, at least, can opt out. More than 230 US airports already use facial recognition technology, according to the website of US Customs and Border Protection (CBP), the federal agency in charge of border security. Governments, meanwhile, claim that facial recognition will reduce the need for paper tickets and make travel more efficient. But it remains flawed and controversial: Early rollouts of facial recognition at airports were plagued by technical and reliability issues, and independent tests cast doubt on the effectiveness of…
US airports are now rolling out facial recognition technology to scan travelers' faces before they board – and Americans, at least, can opt out. More than 230 US airports already use facial recognition technology, according to the website of US Customs and Border Protection (CBP), the federal agency in charge of border security. Governments, meanwhile, claim that facial recognition will reduce the need for paper tickets and make travel more efficient. But it remains flawed and controversial: Early rollouts of facial recognition at airports were plagued by technical and reliability issues, and independent tests cast doubt on the effectiveness of…