Comment Re: A good trade off (Score 2) 65
Those were human made bugs. The point of the article is that AI is finding real actual bugs at a high rate.
Comment Re: Criminal activity in the crapto space. Who car (Score 1) 46
Check your player's handbook, it's definitely an upgrade from studded leather.
Comment Randomization AND a software bug! (Score 3, Insightful) 46
Are you nuts? This is the most interesting news in quite some time, at least from a nerd perspective.
This story has everything you'd want- an insufficient randomization problem, a software bug that makes it subtlety fail and take the wrong path that makes a lot of keys guessable, and then, years later, an eventual exploit where someone, offline, reproduced a ton of these failed key creations, checked the blockchain to see which ones had stuff, set up a burst of activity that would move all the bitcoin at essentially the same time.
It has all kinds of interesting angles- for instance, you can ask if the writers of the software actually were the same as the eventual hackers years later. It's possible that this bug was deliberate, and plenty of cryptocrimes are actually inside jobs. You can also ask why people are so into the idea of hardware wallets in the first place, as using the reference wallet produces a wallet.dat that can be encrypted with veracrypt and stored in a zillion places (and so can the password), and why is that not the first order of business for people looking to store assets with big numbers? Or someone could come in and advertise their own favorite cold wallet instead.
This is a very solid and interesting news story.
Comment Good, fuck those ad peddlers (Score 2) 85
For several years now, any reasonable simple question has had Search-Engine-Optimized useless shit at the top of google- these days it generally can be counted on to fill the entire first page. Google helpfully highlights the relevant text by default so they spend money making sure google can't displau that text but instead displays an ad.
I just googled:
odds of rolling three sixes on three dice
The AI has the 1/216 answer and the reasoning and the math and the links to the people on forums explaining all that. That is actually the search engine right there; it actually links me to people who have explained this to each other over the years.
The top link is to reddit and claims the odds are "180 to where", and eventually has a bunch of links (reddit and google signed an agreement to stop other search engines from reading reddit well). If I click in there I will find the answer. But what if it wasn't?
Under that is stackexchange, and all the wording is chosen so I can't see the answer. Clicking through the answer is present at least. Below that is a dice probability calculator that can make the answer, and something called thoughtco which has the answer in the summary. This is the first non-AI spot on the page with the answer.
Overall the AI summary has been godsent, but if more time has been spent fighting SEO it would not be so needed.
Comment Re:Who cares? (Score 2) 66
The article makes no mention of the source. But everything about this sounds like it's just the CAM release from LuCY. Which has been available for around a week and a half now.
In which case, yeah, the only thing novel about it was that it was posted to a more publicly-visible website.
Nvidia, Microsoft, Meta Warn Against 'Premature Restrictions' of Open-Weight Models 18
Nvidia, Microsoft, Meta, Palantir, and more than 20 other tech companies signed an open letter urging policymakers not to impose "premature restrictions" on open-weight AI models, warning that broad limits could "stifle competition or drive innovation overseas." CNBC reports: They wrote that open-weight models strengthen competition and ensure that the benefits of the technology are "broadly shared rather than concentrated in a few hands." "Relying solely on closed models is not inherently safe: they can be breached, misused, or fail in ways that outsiders cannot detect," the letter said. "And concentrating advanced AI capabilities behind a small number of closed models compounds that risk."
Elon Musk, who runs an AI business under his rocket company SpaceX, also amplified the letter on social media, writing that it has his "full support" in a post on X. SpaceX did not officially sign the letter. Greg Brockman, OpenAI's president, said Thursday that the company believes in broad access, and that he has not been involved in any conversations with the Trump administration about potentially banning Chinese open-weight models in the U.S.
"I think that, that fundamentally, AI and AI usage is something that is actually very important to democratize," Brockman told reporters during a briefing in New York City. "And so, for me, at a sort of deep level, I think that having more models, more usage, that is a good thing." OpenAI CEO Sam Altman addressed the letter in a post on X on Friday, writing that he wants the U.S. to win with both open-weight and proprietary models, and that he is "glad to see this."
[...] In the letter on Friday, the U.S. tech companies said that concerns about unlawful distillation should be addressed through "targeted legal and commercial frameworks" instead of with "sweeping restrictions on techniques that play an important role in AI innovation." "Our AI leadership will be judged not by one frontier AI model, but by whether the United States builds a strong, open ecosystem that diffuses into every sector," the letter said. "This is essential for creating opportunities for innovation and prosperity across the country." The letter follows a separate appeal signed by nearly 200 Silicon Valley companies, including Proton and Y Combinator, warning that restricting U.S. access to Chinese open-weight AI models could cripple the next generation of American startups. "American leadership requires two things: world-leading American open-weight models and continued access for U.S. builders to open models already available worldwide," the startup founders wrote. Instead of broad prohibitions, they argue the government should adopt targeted safeguards.
Of course, these signees "have an obvious economic stake in seeing open AI models flourish," notes TechCrunch. "Companies like Nvidia, Microsoft Azure, and other infrastructure providers have a vested interest in pushing for commoditized models: If models are interchangeable, people will buy more GPUs, rent more cloud capacity, and build more applications."
Elon Musk, who runs an AI business under his rocket company SpaceX, also amplified the letter on social media, writing that it has his "full support" in a post on X. SpaceX did not officially sign the letter. Greg Brockman, OpenAI's president, said Thursday that the company believes in broad access, and that he has not been involved in any conversations with the Trump administration about potentially banning Chinese open-weight models in the U.S.
"I think that, that fundamentally, AI and AI usage is something that is actually very important to democratize," Brockman told reporters during a briefing in New York City. "And so, for me, at a sort of deep level, I think that having more models, more usage, that is a good thing." OpenAI CEO Sam Altman addressed the letter in a post on X on Friday, writing that he wants the U.S. to win with both open-weight and proprietary models, and that he is "glad to see this."
[...] In the letter on Friday, the U.S. tech companies said that concerns about unlawful distillation should be addressed through "targeted legal and commercial frameworks" instead of with "sweeping restrictions on techniques that play an important role in AI innovation." "Our AI leadership will be judged not by one frontier AI model, but by whether the United States builds a strong, open ecosystem that diffuses into every sector," the letter said. "This is essential for creating opportunities for innovation and prosperity across the country." The letter follows a separate appeal signed by nearly 200 Silicon Valley companies, including Proton and Y Combinator, warning that restricting U.S. access to Chinese open-weight AI models could cripple the next generation of American startups. "American leadership requires two things: world-leading American open-weight models and continued access for U.S. builders to open models already available worldwide," the startup founders wrote. Instead of broad prohibitions, they argue the government should adopt targeted safeguards.
Of course, these signees "have an obvious economic stake in seeing open AI models flourish," notes TechCrunch. "Companies like Nvidia, Microsoft Azure, and other infrastructure providers have a vested interest in pushing for commoditized models: If models are interchangeable, people will buy more GPUs, rent more cloud capacity, and build more applications."
Comment Re:Late to the party (Score 1) 49
Xbox 1 emulation is in a so-so place right now. Xemu has pretty good compatibility (80% or so), but it is very, very slow.
To date no one has been able to develop an Xbox emulator using high-level emulation, so Xemu takes a low-level approach based around QEMU. Which makes the computational requirements much, much higher*. To the point where there are plenty of games that run at half-speed (or worse) even on the latest CPUs. The current emulators are heavily CPU limited (and to just a couple of threads at that), and CPUs aren't getting faster very quickly these days.
So if Microsoft has an emulation environment with much lower overhead, then that is a significant development. Because even with third-party emulators available, most games aren't currently fully playable on them.
* LLE extends to the GPU as well. So a lot of cycles are burnt emulating parts of the NV2A GPU and its 25 year old peculiarities
Comment LUENX isn't a knockoff (Score 2) 122
I'm on my third set of LUENX sunglasses. When I inevitably lose this pair (the fate of all my sunglasses), I'll probably buy another. I want purple shiny polarized glasses, they sell those. The say LUENX on the sides. The microfiber cloth that comes with them says LUENX. The little test card for polarization says LUENX (you can only read it through polarized lenses, of course).
So how are these a sketchy brand, or a knockoff, or whatever? These things work fine, arrive at my door the next day, and are like 15 bucks. Am I to believe that these are somehow a bad buy for 15 bucks? Am I to be enraged that I could probably get them for 9 dollars if I went directly to their website, or to alibaba or something?
Anyone have a suggestion for a nice pair of shiny purple polarized sunglasses- ones that aren't Luxotica (aka EssilorLuxottica these days)?
Comment Re:Sojust like every other tech growth story (Score 3, Interesting) 231
Amazon took nine years to reach profitability.
I'm not sure Amazon is a good example here. The company famously opted to reinvest its free cash flow into growing the business, rather than saving them and booking them as net income. They likely could have been profitable sooner otherwise.
Also, I am not aware of Amazon receiving billions in government support in the 1994-2001 timeframe.
Scientists Edited Human Embryo Genes. But Questions Remain (msn.com) 35
"A DNA-editing feat involving editing the genes of early stage embryos was announced this week," reports the Wall Street Journal.
They describe the feat as "a far cry from designer babies, but nevertheless a step in that direction." The research was published online Monday on a preprint server.
They describe the feat as "a far cry from designer babies, but nevertheless a step in that direction." The research was published online Monday on a preprint server.
Comment K'Breel (Score 1) 10
Score one for K'Breel and the Council of Elders, I guess. We'll get you next time, you crafty Martians!
Comment Re:Or switch to Libre (Score 1) 190
Pages, Numbers, and Keynote are still included with every Apple device. The current versions are fully up to date and are free for any device owner.
What Apple has done is put the AI features and licensed stock images behind a paywall. Which isn't great, but those weren't core features to begin with; the iWork apps existed long before they were added.
To quote Apple: "Yes. You can continue using Keynote, Pages, Numbers, and Freeform for free. And while these apps remain free for everyone, an Apple Creator Studio subscription offers premium templates, a library of high-quality, royalty-free photos and graphics, and powerful intelligence features."
Comment Re:For accuracy (Score 1) 190
Why would an otherwise routine software update even come with new EULA terms?
Off the top of my head, I can't recall an Office for Mac update ever coming with a new EULA.
Comment Re:For accuracy (Score 2) 190
While Office 2021 is affected by the expiring license, it's still under support until Oct 2026 and users just need to update. It only reverts to read-only if you don't update.
Thank you. I had been wondering about this precisely because Office 2021 is still receiving updates.
So this is really only an Office 2019 issue. Which still isn't great, but it is at least older.
And from the sounds of things, this only impacts the retail-licensed version of Office 2019. The volume licensed LTSC version doesn't rely on an activation server or certificates.
Comment Re:AI could solve this eventually. (Score 2) 46
It's modded funny because OpenCL is all but dead for new projects. It got weighed down by industry infighting to the point that the big feature of OpenCL 3.0 in 2020 was undoing everything added to the spec after 2011.
So the idea of using OpenCL as a CUDA replacement, rather than something like ROCm or OneAPI, is funny. It's like rewriting C++ programs to use Pascal.
