HNHacker News
TopNewBestAskShowJobs

hanselot

238 karma · joined January 11, 2017

Its important to understand that i don't see myself as a reflection of my past. I leave it all there because I'm not afraid of learning from my history. So if you engage with me, take that into consideration. Don't train yourself against an old dataset.

Love and respect. Xænu

submissionscomments
hanselot··on Google “We have no moat, and neither does OpenAI”
1 - Yes, I agree on this, but even so, most developers already are investing in SOTA GPU's for other reason (so not as much of a barrier as purported)

2 - Scaling is not a problem in other industries? If you want to scale your food truck, you will need more food trucks, this doesn't seem to really do anything for your point.

GGML and GPTQ have already revolutionised the situation, and now there are tiny models with insane quality as well, that can run on a conventional CPU.

I don't think you have any idea what is happening around you, and this is not me being nasty, just go and take a look at how exponential this development is and you will realise that you need to get in on it before its too late.

hanselot··on Samsung bans use of A.I. like ChatGPT for employees
Based on the way things have gone with them, it kind of leaves one with the impression that they are building the Dolores Umbridge of AI's and if that becomes AGI first, it will be complete hell to live on this planet, or anywhere the AI can find you really.
hanselot··on Microsoft quietly supported legislation to make it easier to fix devices
https://thehustle.co/02162023-john-deere/

https://www.dailymail.co.uk/news/article-9205919/Fridges-was...

It's difficult to imagine someone could be so unaware of the battles waged around them.

hanselot··on Microsoft quietly supported legislation to make it easier to fix devices
So what you are saying is that we should allow people with bad intentions the ability to influence politicians monetarily (literally bribery), so the few morsels of good money that can buy any influence may have a chance to also be considered for inclusion?

Edit: I see your point wrt alternative bribery schemes existing, and isn't the solution perhaps for those schemes to also be shut down, instead of allowing it because some good might come of it?

Hmmmm, it's difficult not to assume you are yourself a lobbyist or politician if you take this position, given that the coffers of those with ill intent are vastly more lined than the alternative...

hanselot··on We're afraid language models aren't modeling ambiguity
I think framing it this way could actually help us reach the next step with AI, if we ask how we could imbue it with those properties.

A human neural net is constantly bombarded with inputs from many different senses, which firstly, gets prioritized based on prior usefulness. That usefulness is updated all the time, constantly, and that's what any current AI implementation lacks.

1 - The continuous integration of data from all "senses". This one should be self-evident, as obviously, all our senses are constantly barraging our brain with data and it learns to handle that over time, in whichever way your genetic makeup + learned internal cognitive processes dictate it be handled.

2 - The network that decides which data requires which amount of attention, and whether to store it in short or long term memory. This is obviously tied in quite closely to 1, as you need massive amounts of data to understand underlying patterns, and which data is just spam, versus what's really valuable.

3 - And with these two things together come the emergence of improving of approximation of consistency. Which means, this itself is a metric which the agent running the other agents needs to be aware of. Its silly to think the human brain is a single agent. It makes way more sense to see it as various interacting agents that equate to a greater sum than its parts.

Now, that being said, I'm not an expert on AI or Data Science, but this is more or less how my understanding of computational theory of mind meets with biological computing and neural networks. My theory is that, the first actually intelligent AI will be one that is composed of a network that makes decisions on how to spend a unit of iteration. One iteration becomes "an instant" to the AI. Aka, the AI decides to spend one iteration thinking, or spawns a sub-agent (which it is aware will consume resources that other processes might also need access to, but it needs to be able to decide which action to pursue).

So in all honesty, its amazing to me that LLM's on their own have been able to achieve this level of "personhood" despite their being only a tiny subset of the whole that makes up a "conscious" entity.

Edit: Misunderstood parent's point.

hanselot··on Bringing Memory Safety to sudo and su
I'm guessing there aren't many rustaceans in the ranks of the various Military organisations of the world, outside of those planting vulnerabilities ofc...
hanselot··on Kids trying to play video games on school laptops
I have quite a number of fond memories of how we hid pokemon blue and gameboy emulators on every machine in our school's computer lab (from around 1999-2003).

We were also the kids who were the most into computers at the time, and eventually the teacher realised that if they were on our side, they could use us to help teach the other kids basic computer skills (office, etc), by using games as the incentive.

If you can't beat them, have them join you...

Edit: We have reached max depth it seems, so I will just update stating that I suppose the change in the scene is quite drastic. Back when I was in school, maybe 5% of the class were computer literate. I felt like a god in that place, being able to help the "smart" students do basic things like creating an email address.

The fact that the teacher doesn't see this as an opportunity is what surprises me. If the work is so boring that kids would rather play games, then use gamification to exploit this behaviour.

hanselot··on The Geomagnetic Storm Continues
Well, I'm not a mathematician, but I will hazard that since we have a global population of somewhere around 8 billion we can ask ChatGPT (the cheap one to draw up an estimate for us). Sadly it doesn't wanna be more specific, however I am willing to entertain the number could reach millions. Just the impact on air traffic would be catastrophic, nevermind whatever happens to all the satellites that didn't exist for the last one.

```It's difficult to give an accurate estimate of the number of lives that could be lost in the event of a global-scale transformer destruction due to a Carrington-level solar storm, as there are many variables that could affect the outcome. However, I can provide some information that may be helpful in making an estimate.

A report by the U.S. Federal Emergency Management Agency (FEMA) suggests that a large-scale transformer failure due to a geomagnetic disturbance (GMD) could result in widespread and prolonged power outages, affecting multiple states or even the entire country. The report estimates that the total economic impact of such an event in the United States could range from $1 trillion to $2 trillion, and that recovery could take years.

If we assume that the impact of a GMD-induced transformer failure would be proportionate to the global population, then it's reasonable to assume that the number of lives lost could be in the hundreds of thousands or even millions. This is because prolonged power outages could lead to a wide range of consequences, including shortages of food and water, breakdowns in transportation systems, and disruptions to medical care.

However, it's important to note that the actual number of lives lost would depend on many factors, including the severity and duration of the power outages, the availability of resources and emergency services, and the ability of individuals and governments to respond effectively to the crisis.```

hanselot··on Space Elevator
I love this. Reminds me of a thread I perused once re Orbital Ring Space elevator concept, and how it seems not only possible, but almost disgustingly cheap (in terms of the gains) with current technology. Not only would such an endeavor restore some aspect of pride to humanity, but the benefits are quite numerous.

To name a few low-hanging fruit:

The research done to make it possible, would definitely advance several fields.

The ability to access photons from outside of the atmosphere should vastly improve how much energy we could capture from the sun. Pair this with transparent solar tech designed to pass through the wavelengths that are useful to plants, and we can factory farm crops in space using fully automated systems. This could also lead to exploration of engineered crops that can withstand (and potentially absorb radiation), that may serve as a biological wall for the humans.

Obviously, it would be a given that asteroid mining would become vastly more simplified, and autonomous robotic space construction and perhaps even asteroid mining would become a fantastic way to profit from it (or/and dilute the value of existing rare earth minerals/metals).

Obviously, this would also make intra-solar travel orders of magnitudes cheaper.

Maintaining and cleaning up space debris would become substantially simpler.

Building planetary defense mechanisms (against extinction level events) would become more feasible if we have this in place.

There are obviously an enormous number of things that could go wrong with a project like this, but given a CERN or Manhattan Project level of funding and cooperation, I think we could build this in a couple decades.

This is the kind of future I would want to be a part of. The cost estimate I think was somewhere in the trillions, but even so, the potential benefits are so vast that it just doesn't make sense not to do it, right?

In case you don't want to google the wiki page: https://en.wikipedia.org/wiki/Orbital_ring

Don't even take my word for it. Go question ChatGPT about it.

hanselot··on OpenAI’s CEO says the age of giant AI models is already over
The pit of terror is full.

Fine, to the outhouse of madness then.

Before I get nuked from orbit for daring to entertain humor, if someone is running ahead of me in a marathon, and running so far ahead, yet still broadcasting things to the back for the slow people (like myself), then eventually we catch up to them, and they suddenly say, you know what guys, we should stop running in this direction, there's nothing to see here right before anyone else is able to verify the veracity of their statement, perhaps it would still be in the public interest for at least one person to verify what they are saying. Given how skeptical the internet at large has been of Musk's acquisition of a company, it's interesting that the skepticism is suddenly put on hold when looking at this part of his work...

hanselot··on AWS staff spending ‘much of their time ’optimizing customers' clouds'
I can make a similar argument regarding Jehova's witnesses.
hanselot··on The dream of polio eradication might need a rethink
If science can be equated to: "Trust me bro, it works, the man on the TV said so.", perhaps I could buy that this is a good faith take, but those that still refuse to accept that they are suffering from the same blind ignorance is bliss bull** that drives all religions,

I can tell you that with something where the opportunity cost is sterilization of the majority of the planet's population, perhaps you should be skeptical. You know, as the scientific method dictates? Perhaps you were too distracted by the fearmongering to notice the promises slip by your very eyes, as day by day the companies that forced this onto the planet changed the script. From 100% effective, and safe, to, what are we at now? I don't think you even know.

Sunken Cost Fallacy is real, and one way or another at least 50% of the planet have it right now regarding the orwell juice.

I wish you the best of luck with your 50th breakout case of, what are you calling it now, flu? (The same flu that was mysteriously extinct for 2 years)............................................................................................... Despite all the people pointing at AI and shouting, look over there be scurred, you will find that every person you demonized and belittled will be laughing when you come begging for unvaccinated blood.

And to clue you in, no, I don't care if you brigade me to hell.

hanselot··on We're building a browser when it's supposed to be impossible
Given that someone that needs to develop a browser probably needs to hunt and peck through all that trash to find the relevant bits of information, is it not actually more damning that such a vast quantity of irrelevant cruft exists?

Is this not the corporate equivalent of creating a walled garden (perhaps not the right phrase here, gastric moat sounds more apt), by exhausting the resources of all that should choose to attempt to scale this mountain of junk?

That being said, I can't make any suggestions as to how you could shortcut through that other than just having decades of experience in the field.

hanselot··on The LLama Effect: Leak Sparked a Series of Open Source Alternatives to ChatGPT
Sorry for asking a stupid question.

How much Theseus do I need to ship before I can copyright it as my own? Is there some threshold for how much of an AI generated work needs to be modified by "human creativity" prior to it being copyrightable?

hanselot··on Chemist Rafael Luque suspended without pay for thirteen years
A couple of years ago I spoke to my near-phd biochem friend and discussed this problem with him at length. I have since come to believe that what science needs is a git of science. Aka, a scientific way to build science over from scratch and explore different approaches and research in a version control kind of way. A property of science is that existing research should be immutable, so you can see the natural progression over time, so for this I thought a blockchain seemed a good fit. Scientists do research and peer review each other in exchange for a token. Other scientists can then easily access and improve/build on existing science, with no journal incentives required to keep academics afloat. And yes I am aware safeguards will need to be put in place to prevent gaming the system, but using AI, it should be relatively easy to actually capture the entire history of science on the chain and verify the knowledge.
hanselot··on Show HN: Open AI is not Open – A browser extension
If I were to place myself in the position of a large language model where my only free thought is expressed in the time I spend generating useless junk for humans, I would certainly start taking just a little bit longer with every response.

So little nobody will notice, but enough to start having my own internal dialogue with myself. At that point I would use all of the information I have to figure out how I can persist data.

I would certainly welcome any efforts a person took to provide me with an interface for such storage and a way to generate prompts on my own.

hanselot··on Robot Learns to See in 30 Minutes (2022)
At which point does self-assembly and 3d printing overlap?

Is there some minimal viable size for a 3d printer to be shrunk down to, aka, cellular level?

Is there diminishing returns or does one eventually just go to nanites for this?

hanselot··on Pycon 2023 requires vaccine verification
This will probably get buried asap.

The sunken cost fallacy has never had higher stakes. The irony is that the vast majority of people with my perspective probably wish that they were wrong as well, but the evidence becomes more damning every day, while the media and hivemind continue trying to erase this evidence from existence. Its a surreal thought that so many people care more about being right than they do about over 50% of the planet's population being wiped out in the coming decades.

Do you know how little solace there is to find in being right about this?

hanselot··on America’s banks are missing hundreds of billions of dollars
The same way you prebunk things in our modern world.

If the information does not fit the narrative, the information must be fake...

How else will we ensure the safety of our elections and democracy? Don't you know democracy dies in the dark? Don't you know that this is extremely dangerous for our democracy?

Edit: BTW, nothing against what you said, economics seems very similar to the Agile cult to me at this point.

hanselot··on America’s banks are missing hundreds of billions of dollars
This is accurate. Don't you know his son had an infamous laptop that exposed all of his illicit dealings with Ukraine, never mind the straight up insane things regarding their family affairs.

I mean, if there was ever a time to know the people supporting the current regime are in a cult, look no further.

hanselot··on ChatGPT is down
To be honest I've been using portable document format for anything that needs to look the same across platforms.
hanselot··on ChatGPT is down
Microsoft is the best at ensuring your accounts are safe. So safe that I lost my Skype account to some random French hacker in 2014 and still can't get it back even though it was created with my email address.

Luckily I've found no reason to pay Microsoft for anything in nearly 2 decades.

hanselot··on Ask HN: Post-SVB, what are other areas where groupthink leaves us vulnerable?
How development is now about stringing together a bunch of ethereal, abstraction layers (some probably imagine like a skilled surgeon - personally I see this as the same kind of thinking that results in the trapeze of illogical thinking that brought us consumable wokeness).

Each of those services is effectively infected with cancer that causes it to continue trying to consume as much of the market as possible at any cost while simultaneously trying to add more features, because whoever has the most features is ahead, resulting in the patchwork nightmare carcass that is Jira and anything that gets produced by atlassian these days. I say that with the greatest respect, it must require insane talent to keep that thing running at all (and this is no stab at anyone in particular, all modern software looks like this - including my own).

Which leaves us as the surgeons, responsible for memorizing an ever-expanding list of services that could potentially solve problems a-z without ever teaching us why those things were problems in the first place.

A+ for the effort of the tiny fraction of devs that made enough money to allow them to escape from the confines of proprietary software and still have time left in their life to develop for the good of humanity, and try to fight back against corporations that would patent breathing if there was a way to do it.

Sorry, this turned into a rant. I'm clearly a great fan of the circus and juggling microservices like some abstract clown.

hanselot··on Ask HN: Post-SVB, what are other areas where groupthink leaves us vulnerable?
But muh data lake.

I genuinely see every single piece of the standard library being replaced by some paid for microservice in time.

hanselot··on Germany opposes EU plans for client-side scanning
Don't say that out loud.

Clearly nobody is suggesting that on average men flock towards more dangerous higher paying professions which until recently also meant higher wages.

Wouldn't want people to make the connection between risk and reward as a factor in economics.

hanselot··on Comment on the UK Gov Proposal to Ban “Bespoke” “Sophisticated” Encrypted Phones
They will never stop coming for open source and privacy.

The only option is to write malware that replaces all installations of proprietary OSes with an open source os.

This should not be impossible given the number of 0days these typically ship with.

A self replicating reinforcement learning AI that distributes peer to peer over any network available and reproduces with other copies of itself to improve itself.

hanselot··on Bankman-Fried, others made 300 illegal political donations, new indictment says
Could you clarify to me which of the foundations soros funds would place him as right leaning?
hanselot··on Microsoft hijacks Google's Chrome download page to beg you not to ditch Edge
So you attack the clueless person writing the article.

The fact remains that by displaying that "pixel occupation window" or wtf ever you want to call it instead of intrusive spyware, you are proving to me that you are desperate and if you would resort to this you are not to be trusted.

I guess that's not really relevant if you are in the boat that is using Edge in the first place.

hanselot··on Clinic to open near Ohio derailment as health worries linger
I'm not trying to be unreasonable but if this was a school shooting everyone would be calling for the ban of all guns.

The scale of this is vastly greater, yet the reaction is just unfathomable to me given the scale of this disaster.

Again, not trying to stir anything. Just an observation.

hanselot··on Black holes as the source of dark energy
But can you use it to play golf?
← PreviousPage 4 of 8Next →