HN user
I make things.
I just tried this and it worked without issue.
Some considerations:
1) tell it to extra t the data (in a new session) does that work?
2) if it doesn't, could there be something up with the PDF?
As many commentors suggested, this works well with Gemini so there is likely a missing variable in play.
Share your prompt and the PDF and let's see if we can determine what.
Yes, limiting the full scope of capabilities, which is why I differentiated between Mythos (unrestricted) from guardrailed.
Main point being, we have no idea the measurable capabilities of this, it could be as great or better as unrestricted Mythos, or on par with guardrailed Fable.. or just OpenAI hype that measures up to neither.
The distinction is important because if it truly is a Mythos level (or even guardrailed Fable) it in theory would require that 30 day US government validation before release as well as oversight to the approved partners allowed to use it.
Op was drawing a parallel as to why we should be outraged at the double standard, I was drawing a better parallel by which to compare.
No one commenting on the fact that oAI is releasing a Claude Mythos-class model - with apparent 0 restrictions or concerns by the US government
We don't know that it is Mythos level, it could very well be at (guardrailed) Fable or below.
This is not a wide open distribution, this is only being provided to hand picked partners, similar to how Mythos was distributed (unlike Fable which had wider distribution)
The larger question, which I don't see an answer to in this post:
1) was this tested and validated by the US Government?
2) is the list of partners vetted by the US Government?
If This is "mythos-class" AND
OpenAI approves SK Telecom as a trusted partner ( https://www.wired.com/story/sk-telecom-anthropic-mythos-export-controls/ )
OpenAI did not get approval.
According to Wash they are operationally profitable
https://www.wsj.com/tech/ai/mind-blowing-growth-is-about-to-...
Some question it
https://www.wheresyoured.at/anthropics-profitability-swindle...
Cheap. Fast. Good. Pick two.
Colossus is the world's largest single, unified GPU cluster, all GPUs acting as one coherent supercomputer rather than fragmented pools or multi-site setups. They spun it up in a fraction of the time by all estimates. It's not something you can just throw money at and reproduce the results.
Per Jensen Huang:
"As far as I know, there's only one person in the world who could do that; Elon is singular in his understanding of engineering and construction and large systems and marshaling resources; it's just unbelievable. A supercomputer that you would build would take normally three years to plan and then they deliver the equipment and it takes one year to get it all working."
..."it took 19 days to get Colossus from hardware installation to beginning training, the fastest by far anyone's been able to do that."
https://www.businessinsider.com/jensen-huang-elon-musk-super...
Regarding on site generators. Meta, OpenAI (Microsoft/Oracle) and others are also using on-site gas turbines, generators, and "behind-the-meter" power plants to keep up with the power demand. This has become an industry-wide strategy driven by grid constraints, with natural gas as a fast-deploy option.
It would be great if the grids could keep up with demand, if other options would be considered capable of producing the ongoing demands (ie. more renewable, nuclear, etc) but they're not, and companies are not going to just wait because then they're as good as done.
The answer remains, these rules do not specifically apply to only SpaceX, they apply to a range of companies that fit specific profiles. Timing happens to favor SpaceX, but will equally favor OpenAI, Anthropic and others within the same qualifiers.
The links above provide specifics as to the what's and the why.
Why does SpaceX warrant a change of existing trading rules?
They don't, while timing certainly benefits, and potentially was triggered by them and OpenAI and Anthropic IPOs, these rules are not specific to only apply to SpaceX.
FTSE Russell (Russell 1000/2000 etc.) Adopted "fast entry" for large IPOs. Eligible companies (investable market cap above Russell Top 500 cutoff) can join after 5 trading days (previously quarterly rebalances). Also eased float rules with carve-outs.
https://www.lseg.com/en/media-centre/press-releases/ftse-rus...
Nasdaq (Nasdaq-100): Effective May 1, 2026, top ~40 market-cap companies can enter after 15 trading days (previously 3+ months). Adjusted low-float handling.
https://spotgamma.com/spacex-ipo-index-changes-spotgamma/
S&P Dow Jones (S&P 500): Reducing seasoning from 12 months to 6 months for megacaps and waiving the 4-quarter GAAP profitability requirement for large issuers.
https://www.wsj.com/finance/stocks/stock-indexes-are-contort...
First of all, this is not currently a bomb threat up until someone actually makes a threat.
It makes sense from the perspective of zero tolerance. Any mention or reference is perceived as a threat regardless of additional actions taken.
Tesla has offered insurance on their vehicles since 2019, if they did self-insure the robitaxis, likely they would extend it through their existing insurance program.
For this announcement, no.
Some caveats, this is only for the robotaxi, not any personal owned self driving. Those remain level 2. This is obviously only in Texas.
That said,fFor robotaxis, certainly for accidents that the robotaxi is deemed fully reaponsible, Tesla (or insurance?) will be liable.
Things like other vehicles, third-party interference, passenger misconduct, etc. will share or cover liability depending on the specifics.
Seems like Tesla will self insure, but that's not officially been stated yet.
Some pros for us, in addition to bot protection.
* global distributed caching of content. This reduces the static load on our servers and bandwidth usages to essentially 0, and since it is served at an end point closest to wherever the client is, they get less latency. This includes user logged in specifics as well.
* shared precached common libraries (ie. jquery, etc) for faster client load times
* Offers automated minification of JS, CSS, and HTML, along with image optimization (serve size and resolution of image specific to the device user is viewing it from) to increase speed
* always up mode (even if my server is down for some reason, I can continue to serve static content)
* detailed analytics and reporting on usage / visitors
There are a lot more, but those are a few that come to mind.
Palantir maven uses Claude
The pushback isn't that they use Anthropic, it is that you stated they used it "entirely", which is not true.
Yes Anthropic is a priority model in their ecosystem and they are deeply embedded with both tech and staff, but they are not the one as indicated and sourced in my reply above.
Agreed, which is why a $600 price point on a "budget laptop" targeting users running a web browser seems quite over priced.
Your amazon links are broken.
Thanks. Fixed.
This isn't for people that really even care about performance. It's for people that want a laptop that works with their iPhone
That was my conclusion to my comment in my original. The title of "no other budget laptop can compete" is not just sensationalized, it is factually wrong. It should have been "the least expensive macbook yet comes with a catch"
I’m a bit confused about who this article is really for. The MacBook Neo starts at $600 so when I read:
“MacBook Neo is built on an iPhone chip—the A18 Pro. It’s far less capable of running intensive tasks than any of Apple’s M‑series chips or any moderately powered Intel or AMD processor.”
and that:
“It’s merely the right kind of performance for anybody who wants to browse the internet or stream video.”
...at this price point there are plenty of alternatives for laptops with better performance and specs.
For example, you can get a 15.6" Ryzen 7 5700U laptop with 32GB RAM and a 1TB SSD for less than the “unbeatable” price of the Neo:
https://www.amazon.com/NIAKUN-Computer-Processor-Graphics-Ke...
Or a 15.6" Intel Core i7‑1255U/12650H laptop with 16GB RAM and a 1TB SSD in a similar price range:
https://www.amazon.com/HP-Laptop-High-Performance-i7-1255U-4...
Both of these offer:
* A more traditional laptop CPU
* 2–4× the memory
* 2-4× the storage (1TB vs 256GB base on the Neo)
Standard HDMI/USB‑C video out for external displays
So I can definitely see the appeal of the Neo for people who just want an inexpensive way into macOS, but the claim that “no other budget laptop can compete.” doesn't track.
Maybe it should have been "The least expensive Macbook yet, but that comes with significant downsides."
You should have seen this coming:
https://funeral.star.tupcheck.me/r/oczgRMJWogM
Cause of death: Market saturation with similar tools already exists.
Epitaph: Honesty without nuance is just harshness.
What to build instead: A simple script that generates a one-sentence critique of any given idea.
faq says:
"Audio-to-Sheet-Music: Upload or record audio, get accurate sheet music"
but the bot says:
"I can't process an MP3 for you right now, but I can totally help you generate some awesome classical piano sheet music! Just let me know if you'd like me to create some for you! "
Is this a future feature or is there another way I should be sending audio?
First this is a very editorialized title, the original is:
"Announcing Windows 11 Insider Preview Build 26220.7271 (Dev & Beta Channels)"
Secondly, they're preloading the executable resident in memory to accelerate click to open, similar to what Chrome Browser does on Windows (and websites when browsing)
I don't perceive this as "fixing bad performance, given explorer has never been slow to open for me, but rather further optimizing the experience.
Apple aquires OpenAI, Sam Altman takes over for Tim Cook and brings Jony Ive back.
1) OP indicated his scenario was a self developed app he uses on his own personal device, not a hypothetical "friend". In terms of some unknown future scenario, speculative fear doesn't really provide anything in the ways of a constructive dialog.
2) In regards to inactive accounts, from Google's policy page:
If you have never submitted an app for review and the account is more than one year old, it’s considered inactive.
If you have apps, the account is considered inactive if it is more than one year old, all published apps have fewer than 1,000 combined lifetime installs, the required contact details are not verified, and you have not used Play Console in the last 180 days.
Google sends warning emails at 60, 30, and 7 days before actual closure, allowing time to take corrective actions.
While you are correct that this would lose you access to the developer account, inactivity for a year and ignoring multiple warning messages over a 2 month period gives you an opportunity to weigh your options. It doesn't even require app updates, just activity in the Play console.
OP I was replying to presented his scenario of self developing an app he uses on his own personal device, my response was specifically in regards to that use case, not any hypothetical third party person.
This is why I switched to Android, just for Google now to pull the rug from under my feet again
1) You can continue to install unsigned APKs via adb with the upcoming update.
2) Signing APKs for sideloading requires a Google development account which is a one time fee of $25, no yearly fees.
So still a free sideloading option available, and if you want to avoid adb it is a one time cost that is 1/4 the annual rate on Apple.
for sheets, Gemini exists already natively.
Alternatively, Perplexity Comet browser (or OpenAI Atlas) would presumably provide sidebar functionality to act within your spreadsheets.
Perplexity's response to Reddit's lawsuit:
"Dear Reddit community,
You might’ve read Perplexity was named in a lawsuit filed by Reddit this morning. We know companies usually dodge questions during lawsuits, but we’d rather be up front.
Perplexity believes this is a sad example of what happens when public data becomes a big part of a public company’s business model.
Selling access to training data is an increasingly important revenue stream for Reddit, especially now that model makers are cutting back on deals with Reddit or walking away completely. (A trend Reddit has acknowledged in recent earnings reports).
So, why sue Perplexity? Our guess: it’s about a show of force in Reddit’s training data negotiations with Google and OpenAI. (Perplexity doesn’t train foundation models!)
Here’s where we push back. Reddit told the press we ignored them when they asked about licensing. Untrue. Whenever anyone asks us about content licensing, we explain that Perplexity, as an application-layer company, does not train AI models on content. Never has. So it is impossible for us to sign a license agreement to do so.
A year ago, after explaining this, Reddit insisted we pay anyway, despite lawfully accessing Reddit data. Bowing to strong arm tactics just isn’t how we do business.
What does Perplexity actually do with Reddit content? We summarize Reddit discussions, and we cite Reddit threads in answers, just like people share links to posts here all the time. Perplexity invented citations in AI for two reasons: so that you can verify the accuracy of the AI-generated answers, and so you can follow the citation to learn more and expand your journey of curiosity.
And that’s what people use Perplexity for: journeys of curiosity and learning. When they visit Reddit to read your content it’s because they want to read it, and they read more than they would have from a Google search.
Reddit changed its mind this week on whether they want Perplexity users to find your public content on their journeys of learning. Reddit thinks that’s their right. But it is the opposite of an open internet.
In any case, we won’t be extorted, and we won’t help Reddit extort Google, even if they’re our (huge) competitor. Perplexity will play fair, but we won’t cave. And we won’t let bigger companies use us in shell games.
We’re here to keep helping people pursue wisdom of any kind, cite our sources, and always have more questions than answers. Thanks for reading.
the very first GenAI final footage to appear on screen
Didn't Netflix use "GenAI" for upscaling of old TV shows not that long ago:
https://futurism.com/netflix-ai-upscaling-old-shows-horrific
OP was requesting details of the America party, at the time of writing that was the most detailed list confirmed by Elon Musk in that thread.
Herr is a Time article that provides a summary indicating Musk's confirmation.
https://time.com/7300282/elon-musk-new-political-party-ameri...
As the new party is in its infancy the details beyond that are limited, so it's the best roundup of what "for the 80%" in the middle would imply.
Looks like this was attempted prior as recent as 2022
https://www.theguardian.com/us-news/2022/jul/27/forward-repu...
One caveat is that the intent of this party is not (initially?) to run a presidential candidate, but rather back individual Senators and Representatives.
Backing a candidate for president is not out of the question, but the focus for the next 12 months is on the House and the Senate
https://x.com/tyler__palmer/status/1941615907266232707
reduce debt, responsible spending only
modernize military with ai/robotics
pro tech, accelerate to win in ai
less regulation across board but especially in energy
free speech
pro natalist
centrist policies everywhere else
Obviously that last catch all bullet is where things get complicated.
From Feeds in the sidebar, select Friends.
In my case it was 3 months ago:
latest stats I saw had US at ~53%, followed by 12.9% Mexico, 7.4% United Kingdom, 6.4% France, 6.0% Germany, 4.6% Spain & 9.8% Others.