News for hackers. Stuff that matters.

joshfraser

Re:Things to know if you get cancer in 2026 on Thursday October 15, 2026 @01:17AM (#50099473)
It drives me crazy how many doctors are against elective screenings. When has having more data ever been a bad thing?

Agree that Prenuvo are crazy expensive compared to Mirovita ( https://mirovita.com) and others who offer the exact same service for significantly less.

Re:Things to know if you get cancer in 2026 on Thursday October 15, 2026 @01:09AM (#50099417)
We've all heard someone say "if only we had caught it sooner".

The stats are pretty staggering. Cancer has a 80% survival rate IF you catch it at stage 1. By the time you get to stage 4 your odds drop to 20%. Despite everyone agreeing that early detection is the key, most doctors will discourage preventative screenings because of fear of "causing anxiety" because of the high false positive rate. You're supposed to just wait until you start pissing blood or something.

But you know what's worse than a little anxiety? Getting cancer and not finding out in time.

The best thing you can do is go sign up for a preventative screening with someone like Prenuvo or Mirovita. The price for elective MRIs has fallen drastically over the last few years. I just checked and right now can get a full body MRI with Mirovita for less than a thousand bucks.

I can't think of a better investment for your health and peace of mind.

"experts warn"
You can tie a photo and biometric data to the public key to make it easy to prove that a key was stolen and needs to be rotated.

A better system would have you hold the private key, only share your public key, and sign messages when you want to make payments or approve things.

This is better than the current state of things where we essentially hand out our private keys to random strangers we meet every day (SSN, credit card numbers, etc).

I agree with your sentiment that there are no perfect solutions.

The interesting question is how to verify who someone is before issuing them a digital ID. Estonia verifies people using their Estonian ID cards, their mobile devices, or biometric data if they have it recorded.

The system is only as strong as its weakest link. ID cards can be faked and mobile devices can be stolen. Biometrics can't be easily faked, but they're horrible to have leaked because you can't change your fingerprints or eyeballs if compromised.

KYC = kill your customer

It's time for us to stop pretending that YC checks do anything except provide an illusion of security while putting people's living in danger.

AI makes it trivial to generate fake documents, so most KYC checks can't actually be trusted to verify your identity. As an example of how ridiculous things have gotten, Anthropic launched their verification program for granting access to their Mythos models. North Korea are experts at bypassing KYC checks and were granted early access while the rest of us were locked out.

These leaks are constant and largely unavoidable. Even the largest, most trusted companies in the world get regularly hacked. My passport was leaked and I've received multiple blackmail attempts from people demanding I pay a ransom. There have been multiple kidnappings that have been related to home addresses and private information being leaked.

The situation is really bad, and there are no easy solutions. The correct answer is probably a new government ID system based on public key encryption with some sort of multi-sig between the individual, the government, and your parents (until you're 18). This won't be easy to roll out, but our current system is broken beyond repair. Unfortunately, things probably need to get way worse before anyone cares enough to fix it.

Re:Claude Fable 5.1 and Claude Mythos 5.1 on Tuesday September 01, 2026 @02:27PM (#49525934)
the counterbalance to the AI doomers has always been the fact that everyone has equal access to AI. i hate this new world where Anthropic believe they should be the ones to decide who gets access to super intelligence and who doesn't.
wait, how many workers fall outside the 16-65 range??
Re:The map that keeps Burning Man honest on Thursday May 07, 2026 @03:38PM (#48053834)
Reinforcement training. Camps that leave too much moop don't get invited back.
Bet Claude could build that for you in 5 mins
data centers drive up the cost of power. basic supply and demand.

instead of blocking data centers, we need to scale up energy production. the solution is to get rid of all the red tape that makes it so impossible to build in America.

quality of life metrics are highly correlated to the availability of energy.

Just cancel your subscription. Resubscribe for a month if there's something you really want to watch. You won't miss it.
Re:EFF is leaving X on Thursday April 09, 2026 @04:39PM (#47709640)
We're staying on IG because that's where the queers hang out. What.

I used to be a regular donor to EFF but stopped when it became clear they had lost sight of their mission.

Re:Tesla to pay $243M judgement over Autopilot crash on Tuesday February 24, 2026 @01:11PM (#47140467)
It's wild to see how much people allow their dislike for Elon to impact their critical thinking skills. There aren't some mysterious or unknowable stats. They're being constantly collected and shared with regulators and the general public alike.

https://waymo.com/safety/impact/ https://www.tesla.com/fsd/safety

Re:Tesla to pay $243M judgement over Autopilot crash on Tuesday February 24, 2026 @01:04PM (#47140376)
They do. https://waymo.com/safety/impact/

"90% fewer serious injury crashes for Waymo vs humans"

I said "self-driving software" because I was trying to avoid eliciting a knee-jerk reaction from all you Elon haters.

Re:Tesla to pay $243M judgement over Autopilot crash on Friday February 20, 2026 @03:48PM (#47093731)
For every story like this, there are 10 stories of people who died the old fashioned way behind the wheel.

Accidents like this are obviously tragic, but let's remember that self driving software is already ~10x safer than a human. Unfortunately, lawsuits like this will slow down the rollout of this life-saving technology, resulting in a greater loss of life.

Re:AI Needs More Power on Wednesday December 10, 2025 @07:53AM (#46217145)
I miss the days when Hacker News would root for startups and brave innovators instead of the stogy old incumbents
Re:AI Needs More Power on Wednesday December 10, 2025 @07:45AM (#46217089)
You're telling us that data centers are more sensitive to downtime than airplanes??? That makes no sense.

All of the aeroderivatives were designed in the 70's before we had computer modeling to help optimize the designs. It's not that crazy to assume that we can design a better and more efficient turbine today with all of the help of modern technology.

Re:AI Needs More Power on Wednesday December 10, 2025 @07:40AM (#46217050)
We have an abundance of natural gas and a shortage of electricity.
Re:Hunting for North Korean Fiber Optic Cables on Tuesday December 09, 2025 @02:57AM (#46202409)
What other options do they have? They've been sanctioned to the point where they have few options left but to turn to crime.
"creating software for free that largely benefits large corporations"

Who cares. The end result of this is that we all get to use amazing software, often for free.

Think of your open source contributions as a gift to all of humanity. I wouldn't get too hung up on the fact that bad people can use it. Hammer makers don't add conditions on who can buy their products, even if it could be used as a murder weapon. Take solace in the fact that your work is creating far more good than evil.

You're increasing the rate of innovation in the world. And we're all grateful for it.

Re:Replacing a $3000/mo Heroku bill with a $55/mo server on Tuesday October 21, 2025 @07:00PM (#45662835)
I love the convenience of Heroku but hate their predatory pricing. Who's fixing this?
Re:Find SF parking cops on Tuesday September 23, 2025 @11:42PM (#45356054)
There are a ton of interesting use-cases for public city data. When I was an Airbnb host, I built an early alert system to send me email if my address was ever reported or under investigation. The government moves at a snails pace, so anyone who was paying attention would have plenty of time to cure any issues before any formal investigation was even started. I even had a personal dashboard showing how the enforcement office operated, how many investigators they had, which neighborhoods were getting the most enforcement actions, stats on how cases were resolved, how long they took, etc.
Re:Slack has raised our charges by $195k per year on Thursday September 18, 2025 @10:55AM (#45290482)
I switched from Slack to Discord back in 2017 and I can't imagine ever going back. Their free offering is better than what you get for $$$$ from Slack.

Slack is designed for small groups of people that all know and trust each other. That security model falls apart when you scale to large low-trust organizations. Discord was designed for strangers and offers far more granular controls.

They offer infinite search. Unlimited users. And it's free! Can't recommend it enough.

Modern software is layers upon layers of open-source packages and libraries written by tens of thousands of unrelated engineers. How do you write a spec for that?
Yes. Except our government is the largest buyer.
I've seen the invite-only marketplaces where these exploits are sold. You can buy an exploit to compromise any piece of software or hardware that you can imagine. Many of them go for millions of dollars.

There are known exploits to get root access to every phone or laptop in the world. But researchers won't disclose these to the manufacturers when they can make millions of dollars selling them to governments. Governments won't disclose them because they want to use them to spy on their citizens and foreign adversaries.

The manufacturers prefer to fix these bugs, but aren't usually willing to pay as much as the nation states that are bidding. All they do is drive up the price. Worse, intelligence agencies like the NSA often pressure or incentivize major tech companies to keep zero-days unpatched for exploitation.

It's a really hard problem. There are a bunch of perverse incentives that are putting us all at risk.

It's a classic timing attack. You can detect which Cloudflare datacenter is "closest" (ie. least network latency) to a targeted Signal or Discord user.

The speed of light is the main culprit here.

Re:A New type of web hacking technique: DoubleClickjacking on Friday January 17, 2025 @05:18PM (#42743865)
Back in 2013 I discovered that you could use clickjacking to trick someone into buying anything you wanted from Amazon (assuming they were signed in). It took them almost a year to fix the issue. They never paid me a bounty.

https://onlineaspect.com/2014/06/06/clickjacking-amazon-com/

Older »