It feels hard to be excited by a DGX Spark stuffed into a laptop. It's still slow RAM (much slower than a Mac) and arbitrarily limited to 128GB. Can't they at least offer a higher end model with faster RAM and more of it? Sure it would cost a bundle, but there are still people who'd buy it for the local AI capability.
Per the article, “much of the original documentation has been lost or fragmented. Voyager paperwork from the 1970s and 1980s was largely paper, and each time the project moved offices, more of it disappeared”.
Maybe not years ago, but scanning documents with the phone in your pocket has become incredibly efficient. That combined with AI transcription and indexing for search makes such a project faster and cheaper in 2026 than at almost any other time in the past.
I scanned an entire book in 2016. It was way faster than scanning with the phone in my pocket. It’s just not ergonomic to use a general purpose device (a phone) instead of a specialized device (dedicated scanner).
Agreed. I scanned a short book with my phone, and a dedicated scanner would have been nice to have.
But with page flattening and separation and automated capture, it went much faster than I would have thought. If I were going to do a lot more, I'd want something like a scan tent [1]. It's not as ergonomic as a dedicated solution, but in 2026 a phone and some light can get you a lot of the way there, pretty fast.
I mean, they were never meant to last this long. NASA has a shoestring budget. I understand not taking the time and resources to do that when it could stop working a week later.
Really glad to see someone asking this question. After building a fairly significant AI tool using Python tools, I really wish AI/ML tools would all be rewritten to use an actually performant language - say, Rust - without transitive dependency hell on all the package versions.
The vast majority of Python's AI/ML ecosystem is already written in C/C++ and uses interop glue to call it from Python. But agreed on the transitive dependencies, it's a nightmare
Well, when there's a software created accident, possibly horrendous, I'm sure Tesla will be relieved that it's one case where they can point to it being a completely unauthorized system for which they reject responsibility.
Because of almost certainly fraudulent science. See https://www.science.org/content/article/potential-fabricatio.... Research was misled for decades down the path of assuming that Amyloid plaques are causative of Alzheimers, a now disproven theory (there's an association, but that's not the same as causative.)
Scientist Ruth Itzhaki spent years studying a far more promising theory of Alzheimer's: that it's caused by viral infection in the brain, particularly HSV-1, best known for causing cold sores. Most have it, so there are clearly other factors at work, likely related to susceptibility in particular individuals to to the virus infecting the brain and spreading over time. See https://pubmed.ncbi.nlm.nih.gov/34205498/
The implication is that anti-viral treatments are likely to inhibit and potentially cure Alzheimer's. There is already unintended evidence along these lines, both via antiviral drugs and vaccines.
I see several people here who appear to be replying to the headline rather than actually reading the article (well, podcast transcript, technically). What you mention is already discussed extensively in the podcast episode, and Matt Shrag (the scientist your first link is about) is the one being interviewed in the podcast.
reply