News for hackers. Stuff that matters.

AbstractH24

Re:Early progress on accelerating drug development with Claude on Saturday October 31, 2026 @03:34PM (#50319317)
I yearn for the day when Anthropic moves on to dealing with big problems and leaves the mundane to cheaper models that focus on volume over quality, because it's no longer worth their time.

It'll mean the AI we all need in daily life is nowhere near the frontier anymore.

Re:AI promised less work. I've never worked more on Saturday October 31, 2026 @03:31PM (#50319287)
The question becomes, are you more productive than ever? I (or my agent) can do a lot of interacting with GitHub and nothing of value can come of it.

If more was expected of everyone and everyone was producing more, it'd be a different issue.

Re:AI promised less work. I've never worked more on Saturday October 31, 2026 @03:28PM (#50319263)
Well put
At what point will this exemption become an unreasonable request?

Some exemptions businesses are required to accommodate, like putting in a ramp to an office so someone can get to a desk. Others are considered unreasonable, making it possible for someone who uses a wheelchair to work on high-tension wires.

Re:It's now official, tech is the new tobacco on Saturday October 31, 2026 @03:16PM (#50319145)
I think the time for nerds to shine has already come, and maybe past.

The fact this headline conflates tech and social media is evidence. Disatisfaction with the way AI, social media, and inequality has changed life is all being packaged up and called "tech."

Re:Tell HN: Scribd Breached on Saturday October 31, 2026 @01:30PM (#50318137)
Tumblr too?
Re:Tell HN: Scribd Breached on Saturday October 31, 2026 @01:29PM (#50318132)
What actually is Scribd?

They were always just that thing that showed up in Google searches cause they scraped presentations and stuff that was unknowingly public. But I didn’t quite know why anyone found them of value.

Will be interesting to see if November 2026 is looked at as the inflection point or a hiccup for markets.

Ask HN: Is AI destroying public online community as a protective moat?

Initially my question was going to be about Reddit and how/why they let it go down the tubes with AI slop and declining confidence in ranking to a human. But as I think through it, the question applies to all of social media. Are they afraid if they meaningfully prevent bots traffic and [perceived] shareholder value will drop?Counter argument is X/Twitter. Musk has contorted it in all sorts of...

( Read More... | no comments | 2 points | on HN )
The silent majority doesn’t live on Hacker News
Re:Woman with epilepsy dies after taking ChatGPT medical advice on Saturday October 31, 2026 @08:54AM (#50315759)
What this, and the article, really point to is a need for folks in tech and other high performing areas despite their epilepsy to come together.

Other groups I’ve found in recent years have provided support and shown me for the first time I wasn’t alone. But none talk a lot about this.

Re:Claude 20x is now Ultra without the 5-hour limit on Saturday October 31, 2026 @08:51AM (#50315734)
Nope
Re:US tech stocks see sharp falls over AI cost fears on Saturday October 31, 2026 @08:49AM (#50315718)
Interesting enfough, wave years tend not to produce much of this post-election bump. And often a small near term dip followed by a muted year

But let’s consider what causes a wave. Economic anger.

Re:US tech stocks see sharp falls over AI cost fears on Saturday October 31, 2026 @08:44AM (#50315676)
Th difference now is Trump will be on his way out
Re:US tech stocks see sharp falls over AI cost fears on Saturday October 31, 2026 @08:41AM (#50315650)
It when will the past become good enough for daily use?

They make new car models each year. Plenty of folks drive old ones. Few do annual upgrades.

How typical is this of tech ipos? Or ipos in general?

I realize that SpaceX only did slightly better. But not sure what that says

Re:Claude 20x is now Ultra without the 5-hour limit on Friday October 30, 2026 @11:16PM (#50312894)
Given that it’s unclear what x is multiplying, isn’t it all slight of hand?
I haven't seen ads on google search in like 6 years because of blockers

Easy to forget that others still do.

Re:Woman with epilepsy dies after taking ChatGPT medical advice on Thursday October 29, 2026 @01:18AM (#50284834)
Well put

-another with weird brain waves.

Re:Brown student maps the hidden global labor powering AI on Tuesday October 27, 2026 @02:32PM (#50263679)
>To isolate workers to make them powerless gives even more power to companies that are already very powerful.

How far we've come from everyone saying "WFH or we quit, its our right!"

Re:Brown student maps the hidden global labor powering AI on Tuesday October 27, 2026 @02:31PM (#50263667)
> Cool project bad political motivated wordings in the post.

Sad part is the kid likley doesn't realize that.

The fact this doesn't seem like a bigger deal shows how numb we've all gotten to these.

A quarter of their workforce! While a major conference was going on!

Re:Neural chip decodes interactions between brain regions on Sunday October 25, 2026 @10:49PM (#50238491)
The kind of keto needed to actually impact epilepsy is so strict it’s nearly impossible to stick to it and live life. Particularly as an adult.

If you eat an apple you’ve eaten all the carbs you can have in a day.

That said, putting epilepsy aside I clearly function better as a human when I have less carbs. And I’m interested in these wearables coming to market [finally] that track ketones along with blood sugar.

Re:Neural chip decodes interactions between brain regions on Saturday October 24, 2026 @08:56PM (#50227107)
Thanks for the wishes. I'm super lucky for the time I live in and access to cutting-edge medical care.

Other than a slightly worse-than-normal memory, I live a pretty normal life. Most people don't realize even I have epilepsy, let alone a computer in my skull with wires running deep into my brain sending electric pulses.

If I didn't live when I did and where I do (NYC) my life would be very different. And relative to many others, my seizures are not so bad.

Re:Neural chip decodes interactions between brain regions on Saturday October 24, 2026 @08:54PM (#50227095)
It's been 4 years since I got it, and been quite a journey. I'll do my best to try and answer the question succinctly:

The RNS succeeded as a tool for capturing orders of magnitude more detailed data about how my brain functions and how medication impacts it. It also had some success in reducing the the severity of the fairly infrequent clinical seizure I do have. But, not only did it fail to live up to its promise of getting me off medication, it didn't even prevent me from needing to start taking one that's effective but has some pretty strong side effects (Xcopri).

Titrating onto Xcopri is a long, slow process. I started over the summer, and my doctor could see immediately see the degrees at which subclinical seizure-like activity stopped. Then, as the titration continued and the side effects started to become noticeable, my doctor could see how adjusting my other meds to reduce side effects impacted activity and could do so comfortably with a much finer tooth comb than he would otherwise.

Without the information the RNS provides, I would likely be unable to tolerate the Xcopri and the next step would be another brain surgery (although without the RNS, my epilepsy would appear to be much better controlled than it is. Because I rarely have a clinical seizure.)

So it seems to me that while these implanted devices have real power in their ability to collect data, their ability to reliably alter the brain function is questionable. But that data itself is huge . Hope you keep working in the space!

(glad to answer more questions; feel free to ask or DM me. I'm a bit more informed about this than the average bear)

Re:Neural chip decodes interactions between brain regions on Saturday October 24, 2026 @05:48PM (#50225704)
Most of this already exists, and I already have it implanted in my brain

https://www.neuropace.com/

Articles can be weird sometimes in the details they include/omit.

Re:No Signs of AI Payoff Outside the Tech Sector on Saturday October 24, 2026 @04:14PM (#50224726)
There is two ways of looking at this - one is that things are overhyped.

The other is that there is so much more value to be extracted before this bubble burst.

Not sure which I believe, objectively. Subjectively, inside of tech, it certainly feels like the former.

Re:The Other Epidemic: Trying to get off benzodiazepines on Wednesday October 21, 2026 @11:15AM (#50182446)
For just shy of 5 years I was prescribed and took a benzo daily for my seizures called Clobazam. It's relatively weak, extremely long-acting, and slightly structurally different than the ones people most frequently abuse.

Today marks a week since I've totally stopped. It's been interesting to see the relatively low-stakes rollercoaster that getting off it has been. The bigger roller coaster was how sedated I was last month, because before stopping I started another epiplepsy drug that increased how much of this benzo I metabolized. While I see how this experience has ranged from euphoria to insomnia to exhaustion, it's certainly been nothing like this author (so far).

When I started this benzo, my psych said something to me I'll never forget: "oh, there's nothing wrong with taking benzos daily... as long as you take them every day at exactly the same time for the rest of your life." My epileptologist swore that wasn't the case and seems to have been right, but that advice still lingers with me about benzos in general.

When I learned video blackjack is playing the odds in a way that tickled my brain I’ll admit I enjoyed it. I’m sure for some prediction markets do the same. If you can stay level headed there’s probably easy money to be made on prediction markets trading people’s momentum. Staying level headed isn’t easy.

Bummer that tables on the strip in Vegas have such high minimums. Couldn’t even get myself to play one hand on them (but losing $500 in a matter of minutes on Kalshi or Polymarket seemed like nothing)

In the past 5 years I’ve been to Vegas & Atlantic City 7 times for conferences and music festivals. Never had any interest in gambling, except $20 in video blackjack on the most recent trip.

At the end of the summer I dipped my toe in Polymarket while watching the US Open (the only sporting event that interest me). In a week and I half I nearly lost 2 months salary. Found myself up at 3AM betting not just on tennis, but also on random sports that seemed like a “sure thing.” And not just on Polymarket, but Kalshi too.

The dark patterns involved in prediction markets are real and terrifying. Washed my hands of that, but woof.

Older »