You Can Now Run a GPT-3-Level AI Model on Your Laptop, Phone, and Raspberry Pi
Benj Edwards | Ars Technica
“On Friday, a software developer named Georgi Gerganov created a tool called “llama.cpp” that can run a new GPT-3-class AI model of a large language, LLaMA, locally on a Mac laptop. Shortly after that, people also learned how to run LLaMA on Windows.Then someone showed it running on a Pixel 6 phone, and a Raspberry Pi came next (albeit very slowly). If this continues, we could be looking at a pocket-sized competitor to ChatGPT before we know it.”
A Gene Therapy Cure for Sickle Cell is on the Horizon
Emily Mullin | Wired
“[Evie] Junior…is one of dozens of sickle cell patients in the US and Europe who have received gene therapies in clinical trials—some led by universities, others by biotech companies. Two such therapies, one from Bluebird Bio and another from CRISPR Therapeutics and Vertex Pharmaceuticals, are the closest to come to market. The companies are now seeking regulatory approval in the US and Europe. If successful, more patients may soon benefit from these therapies, although access and affordability may limit who can get them.”
This Couple Just Got Married in the Taco Bell Metaverse
Tanya Basu | MIT Technology Review
“The chapel at the company’s Taco Bell Cantina restaurant in Las Vegas has married 800 couples so far. There are copycat virtual weddings too. ‘T Bell saw the brand’s fans interacting in the metaverse and decided to meet them literally where they were,’ said a spokesperson. That means dancing hot sauce packets, a Taco Bell-themed dance floor, a turban for Mohnot, and the famous bell branding everywhere.”
Inside the Global Race to Turn Water into Fuel
Max Bearak | The New York Times
“A consortium of energy companies led by BP plans to occupy an expanse of land eight times the size of New York City with as many as 1,743 wind turbines, each nearly as tall as the Empire State Building State Building, with its 10 million or so solar panels and more than a thousand miles of roads to connect it all. But none of the 26 gigawatts of energy the site is expected to produce, equivalent to a third of currently required by Australia’s grid, will go to public use. Instead, it will be used to produce a novel type of industrial fuel: green hydrogen.”
Has the 3D Printing Revolution Arrived?
Tim Lewis | The watchman
“i‘What happened 10 years ago, when there was this huge hype, there was so much nonsense written: “You’ll print anything with these machines! It’s going to take over the world!”‘ said Hague. ‘But this is becoming a really mature technology, it’s no longer an emerging technology. It’s being widely implemented by the likes of Rolls-Royce and General Electric, and we’re working with AstraZeneca, GSK, a whole bunch of different people. The printing things at home will never happen, but it has developed into a multibillion-dollar industry.’i”
AI-Imager Midjourney v5 Stuns With Photorealistic Images—and 5-Fingered Hands
Benj Edwards | Ars Technica
“Midjourney v5 is now available as an alpha test for customers who subscribe to the Midjourney service, available through Discord. ‘MJ v5 currently makes me feel like I’m finally getting glasses after ignoring bad eyesight for quite some time,’ says Julie Wieland, a graphic designer who frequently shares her Midjourney creations on Twitter. ‘Suddenly you see everything in 4k, it feels strange but also surprising.’i”
MANAGEMENT
AI Generated Images From Text Are Not Copyrightable, US Government Rules
Chris Holt | Engadget
“That’s according to the US Copyright Office (USCO), which likens such prompts to a consumer giving directions to a commissioned artist. ‘They determine what the prompter wants to depict, but the machine determines how to implement the those instructions in its output,’ the USCO wrote here published in the Federal Register. ‘When an AI technology simply receives a prompt from a human and produces complex written, visual, or musical works in response , the “traditional elements of authorship” are determined and enforced by the technology—not the human user,’ the office said.”
GPT-4 Has the Memory of a Goldfish
Jacob Stern | The Atlantic
“At this point, the many flaws of AI-based language models have been analyzed to death—their inexorable fallibility, their capacity for bias and bigotry, their lack of common sense. …But the big language models have another shortcoming that has so far been overlooked: their poor recall. These multibillion-dollar programs, which require several city blocks’ worth of energy to run, can now can code websites, plan vacations, and draft company-wide emails in the style of William Faulkner. But they have the memory of a goldfish.
Microsoft Removes an Ethical AI Team as It Doubles Down on OpenAI
Rebecca Bellan | TechCrunch
“The move calls into question Microsoft’s commitment to ensuring that its product design and AI principles are closely aligned at a time when the company is making controversial AI tools available to the mainstream. Still maintaining Microsoft’s Office of Responsible AI (ORA), which sets the rules for responsible AI through governance and public policy work. But Platformer employees said the ethics and society team is responsible for ensuring that the Microsoft’s responsible AI principles are actually reflected in the design of the products being shipped.”
It’s official: No more Crispr Baby—for Now
Grace Brown | Wired
“After several days of experts chewing over the scientific, ethical, and governance issues associated with human genome editing, the [Third International Summit on Human Genome Editing’s] the organizing committee issued a final statement. Hereditary editing of the human genome—the editing of embryos that are then implanted to establish a pregnancy, which can pass on their edited DNA—’remains unacceptable at this time,’ the conclusion of the committee. ‘Public discussions and policy debates continue and are important for deciding whether this technology should be used.’i”
Photo Credit: Kenan Alboshi / Unsplash