I think ads were better because everyone was aware of the conflict of interest issues and took steps to manage it. They admitted it could be an issue and would have internal firewalls where ad sales weren't allowed to talk to the journalists, things like that.
The problem with the "philanthropic" funding is not only is there no firewall, but it's explicitly pay-to-play agenda pushing. So that culture of keeping revenue and news separated has vanished without the public being made aware of it.
The right wing are saying things like, journalism is fake because they're just pushing a left wing agenda for money. That's the claim that delegitimizes journalism. But it turns out that the rightists are correct and that's exactly what journalists are doing, so it's hardly their fault that mistrust was created and cultivated. You should mistrust people who mislead you about their agenda, and if you go to the AP website today and look at some random article paid for by these grants there's no mention of where the funding comes from. Actually you get a popup begging for donations, saying that because the AP is a non-profit it's extra trustworthy.
If they want to regain trust, a good start would be by refusing to accept money for stories, relabelling all the stories where they did that, and then pay for journalism purely out of subscription revenues or ads brokered through exchanges (i.e. where the advertiser doesn't have any contact with the publishers).
Relevant builds a data platform that helps safety-net healthcare providers deliver better care using data from their electronic health records. Our customers are non-profit clinics called Community Health Centers. You can read more about them, us, and the role on our jobs page: https://relevant.healthcare/jobs/
We're a small team (~25 overall). We're bootstrapped (no external capital whatsoever, VC or otherwise) and profitable. We've got a modern tech stack (Rails, Typescript, React, GraphQL, d3) and a healthy approach to both tech debt and work/life balance… and we’re also a very nice group of people that get to work on interesting problems for some of the few unimpeachably good actors in healthcare!
Looks like all three breaking changes are patches for inadvertent chain of thought disclosure. Someone found out (don't have the tweet handy) that if you created a bogus "think_deeply" tool and then forced the model to use it, it would output what is believed to be its raw thinking there - I believe the first breaking change stops this. The second two are aimed at people getting Haiku to repeat thinking blocks from other models verbatim. I get that in their eyes it's an "exploit" but still kinda disappointing that they patched this
You claim not to seek approval, but you post little smiley faces at the end of every post, which tells me you desperately want my attention. If you didn't care how people receive your opinion, you wouldn't be conspicuously trying to convince the same people you don't care how they receive your opinion.
Arguing with an indifferent brick wall is not what a fella would consider resilient or emotionally mature behavior. Frankly, any halfway competent bonafide troll would get a kick out of knowing he or she played you into wasting precious time out of your own life.
(1) the before and free images are pretty small and limited to just parts of the frame, like one image had a plant in it that had finer detail with DLSS but the rest of the frame looked the same. If they were serious about selling the thing they could have found better demo images.
(2) It’s a chronic problem that NVIDIA is coming out with new features that only work with the latest chips, right up there with Intel’s inability to deliver accelerator benefits across the ecosystem. Back in the pandemic when graphics cards were impossible to get it was just cruel, we are heading into similar territory with the AI bubble. Like I can see the motivation to get people to upgrade to the latest and greatest but in a shortage like we have now it is just brand destruction.
He really doesn't want to think about it anymore, but I'm sure he's not thought through the potential for the Epstein-Trump files to become a story again in the media without the Iran war to preoccupy the pundits
I agree and wonder whether its either people who basically never use the model complaining or people who used it once a long time ago and haven't touched it since.
We have access to Fable at our company on our enterprise plans and most of us rarely run into an issue.
Obviously this is gonna vary a lot with what technical domain you work in which is why its important when talking about the classifiers that people specify exactly what types of workloads they were seeing failures with.
The whole point of avoiding the verification in my case is for privacy reasons... I don't want google or anyone else tracking what I do through the use of an account.
> Or you could go through the android phone sign-up process which doesn't require one
This is worse IMO because now the number is associated with that device forever. And unless I'm willing to risk my account to compromise from a future owner of the same number OR device, I must now keep both... forever.
> grey web
I don't want to give them my info either, nor have my account associated with sketchy individuals.
> you could become a phone company
I do own DID blocks but this is unhelpful because google's verification specifically requires SMS over real mobile numbers, and I'm not interested in becoming an MVNO or cellular carrier.
LTO-9 increased track density, increased the total number of tracks by reducing their width and increased linear density.
The "how" certainly does not sound "much simpler" to me ?[1]
The new, narrower track width was achieved through a combination of improvements in tracking
performance, new, narrower tunneling magnetoresistive (TMR) readers within the head assembly
(now <1000nm in width) and optimized writers. The narrower tracks also necessitated
improvements to the media magnetic layer. Magnetic particles are Barium Ferrite (BaFe) as in LTO-8, but with
improved characteristics to support the higher linear densities.
Changes were also made to the tape substrate, with the thickness reduced by 7% from 5.6µm to
5.2 µm. This in turn enables more tape to fit on the reel – tape length per cartridge has
increased from 960m to 1035m. The new substrate also features optimized stability
characteristics to mitigate tape dimensional variations that occur due to fluctuations in
environmental conditions.
In addition the LTO Program's own technical paper describes how LTO-9 manages dimensional changes in the new substrate with a one-time-per-tape calibration algorithm to control 32-channel recording via closed-loop servo algorithms, plus a longer-block Reed-Solomon C2 code — together getting LTO-9's uncorrectable bit error rate to 10⁻²⁰, a 10× improvement over LTO-8[2]
In the US there is ACH transfers which are free and slow and reversible and wire transfers which are expensive and fast and normally irreversible. Fednow is a new fast payment system.
And because it's accurate the AI will know how to weigh that accurate information against other accurate information in such a way that it will give you a great answer without any help. In other words: any amateur will be able to do aircraft maintenance! What can go wrong? /s
I've used Fable for so much stuff. My experience has been that it can pretty much one-shot most of my complex problems, if I describe them clearly and provide a solid way for it to verify its work.
I get punted down to Opus 5 occasionally (for security-adjacent things) but that's pretty rare.
Just remove the open collective link. Point to your homepage where the motivated donors will certainly pay. You already should know how many of the motivated users actually care about donations. The goal of this story and publishing it in HN seems to catch more eyeballs and marketing of Ankidroid, under new leadership. All I am suggesting is, just play along with Google. No harm here.
I don't mean to sound like I'm dismissing your experience, but are you sure? I've (semi regularly, most of the time I'm even trying to use Fable) started with Fable, proceeded through my planning, and then at some point in the future realized it had kicked me back to Opus without me knowing. It obviously _said_ it had happened, but I didn't realize and just continued. This might primarily be a result of the project I'm working on (anything network related seems to gets kicked back).
I'd guesstimate that ~80% of the time I thought I was using Fable, I wasn't actually. It's also led me to just... not even try, and just start with Opus regardless.
I've found Fable unusable; not because it's bad, but because it... can't be used.
You can easily trip it up if you're doing reverse engineering work. From memory, the moment Fable 5 saw anything loosely related to "linux seccomp" it threw a fit.
I have no doubt that in the future AI code will be superior to human code. But today is not that day, and these projects that have never yet had human eyes on them are mostly shoddy pieces of work.
I've worked with a number of clients that have lost senior engineers due to old age or retirement, and the impact is often very significant regardless of how sophisticated their knowledge management infrastructure is. Documentation can not fully replace the insight and situational awareness many of these experts posses and most of the time knowledge management still suffers from the 'not knowing what you don't know' issue. Only well structured 'apprenticeship' programs come close. I know some are going to point to AI, but I've worked in AI for over 2 decades and have extensive experience in Expert Systems and Knowledge Engineering from early in my career. It is not a complete solution. Given today's technology its trivial to replicate 'text book knowledge' but the difficult, expensive problems occur in the margins and its still hard to train AI on open ended / rare occurrences outside of domains that have specific advantages like software engineering. Talented and experienced people are still required. Maybe that will change soon, but AI isn't there yet.
“ Claude Fable 5.1's writing is generally a step up from earlier Claude models, with fewer stock phrases and less unexplained jargon. In some cases, though, its prose is denser than Claude Fable 5's: sentences run longer and there are fewer paragraph breaks.”
I cancelled my pro max Claude subscription last week; codex is much more succinct. I am curious if this is getting better.
I don’t think Anthropic realizes that humans have a token limit too and it can be exhausting to read Claude’s output. Prose density is not the same thing as succinctness.
