2,284 karma · joined October 4, 2016
My personal views, not that of my employer or any other affiliations.
So many startups trying to automate sales, but somehow the two biggest frontier labs have decided that the best GTM strategy is firmly human-in-the-loop.
For a pretty funny comment about pricing.
https://www.reddit.com/r/chipdesign/comments/1ajrli2/cadence...
IMO, the fundamental issue for preventative screening is there is basically no amount of money I would not part with (of my money, the insurer's money, or private debt) to not die. I expect this is true for most people, and it makes preventative screening a tricky topic. In recommending screening for those >x age, you will miss some detectable, preventable and treatable cancer risk for those <x age, purely for cost. No one wants to be explicit about that though!
I think the only way out of that uncomfortable conversation is making screening so cheap via automation that you can basically run it for very low incremental cost as often as individual risk tolerance permits. This would be paid for on the back of earlier interventions vs late-stage, expensive interventions.
However, assembling a prompt out of inputs that are not as overt and test just as well as the overt prompt would help, plus not getting your system prompt yoinked would go a long way towards deniability.
DARPA has probably been going after this since Attention is all you need.
So you test it like a black box, but IMO that suffers from the same pollution any of the other tests (coding ability, math ability, w/e) that currently suffer from, except it's even harder to evaluate objectively.
I don't think so. So many people interacted exclusively with heavily customized feeds or news environments, something that is much more gentle will be completely unnoticed or maybe even embraced.
> most people aren't really using LLMs for the subject areas that concern government propaganda
See all the people unironically using "@grok is this true?" It doesn't have to just be government propaganda (eg did Nixon break into Watergate?), it is more about shaping the boundaries of a conversation, framing, etc.
> You seem to be envisioning some kind of a world where people don't access the news or social media directly, but it is somehow passed through some kind of LLM transformation filter.
I envision a world where most people take the path of least resistance. They will not explicitly sign up for it, but will gradually shift to reading the easily digested stuff first. Look how popular tiktok is, the popularity of summarized info, etc. In that summarization and aggregation, there is plenty of room to steer a conversation or influence thought, especially over a large audience.
There is nothing here that will be an overt smoking gun, just a systematic bias towards a particular idea, thought, etc. Hard to prove and even harder to know it's happening.
It has never been cheaper or easier to influence millions of people, either deniably-subtly (though omission, selective results, "hallucinations" etc) or via sock puppetting.
If I am a government, there is nothing more valuable to me than being able to control the discussion, the overton window, and the prevailing narratives. LLMs are a very low cost way to do that, can be tailored at the individual level (unlike most current TV news, personal "feeds" etc) and have the benefit of a huge volume of context.
The models are effectively black-box weights and are resistant to bias-tests. IMO, a key development will be having an "overlay" of weights to apply on top of a "clean" world model that is tailored to whatever interests can pay for it. Being able to serve that overlay dynamically, or atleast per-user is the killer app.
split deployments -- perhaps you want to see how an update impacts something: if error rates change, if conversion rates change, w/e. K8s makes this pretty easy to do via something like a canary or blue green deployment. Likewise, if you need to rollback, you can do this easily as well from a known good image.
Perhaps you need multiple servers -- not for scale -- but to be closer to your users geographically. 1 server in each of -5-10 AZs makes the updates a bit more complicated, especially if you need to do something like a db schema update.
Perhaps your traffic is lumpy and peaks during specific times of the year. Instead of provisioning a bigger VM during these times, your would prefer to scale horizontally automatically. Likewise, depending on the predictable-ness of the distribution of traffic, running a larger machine all the time might be very expensive for only the occasional burst of traffic.
To be very clear, you can do all of this without k8s. The question is, is it easier to do it with or without? IMO, it is a personal decision, and k8s makes a lot of sense to me. If it doesn't make a ton of sense for your app, don't use it.
Spend some time learning it, using it to deploy simple apps, and you won't go back to deploying in a VM again imo.
This only gets better with ai-assisted development, any model is going to produce much better results for k8s given the huge training set vs someone's bespoke build rube-goldberg machine.
However, the tech exists for a reason and is not inherently bad, the issue is the lock-in, the lack of choice and interoperability.
IMO, there is plenty of space for an OEM who can play nice with others, offer an open (and vibrant ecosystem), and keep users coming back by choice, not by lock-in.
Idk about this, I have gotten almost every job I have ever had on cold-apply, including internships. The only one that wasn't that way was talking to a (internal) recruiter in college.
Don't discount that path. I did not have the best grades or anything, but (IMO) a mix of skills that was a good fit for the job at hand and confidence I could apply them.
Most of the people you will interact with in the (corporate) world have no understanding of their own understanding, and are operating in unknown unknown territory. Being confident, demonstrating competence in something jointly known/unknown or known/known helps a ton.
I don't even work with it that much and have a laundry list of complaints about the weird little edge cases or funky pieces of documentation required to make things work.
There are a lot of alternative cements to portland, interested to see if that is in-scope. The list of admixtures is also very long and also fairly secretive. UHPC is a pretty cool development, and I am especially bullish about removing rebar and replacing it with FRP bar to limit the eventual rust cracking that comes with the gradual march of carbonation.
Anyways, very cool and looking forward to the mix developments that come out of this framework.
At least in the US, you have 70+ year old lawmakers proposing (not even writing) laws they do not understand, passed to them by opaque groups with a obscured, albeit clear, interest.
See the latest age verification bills passed by Meta through a convoluted web of influence. Bring back the technocrats.
To answer the point, there is no technical limitation keeping people in the UK from building, creating and shooting homemade or otherwise improvised guns that I am aware of.
What the UK does have is universal healthcare, a 3-4x lower incarceration rate and dramatically improved social safety services.
I think you can group the majority of shooters into three buckets -- ideologically driven (think white supremacists, Islamic terrorists, anarchists, etc), the mentally ill, and the criminally motivated (gang shootings mostly). The US has only amplifying factors for all three groups.
For idealgoues, there is no wider span of acceptable discourse than in the US. Commonly espoused views in the US legislative and executive branches are criminal offenses in a number of peer countries, e.g hate speech is still constitutionally protected speech in the US. The rhetoric is insane, accusations of nazism, faciscm from the left and similar accusations from the right, and generally a very high degree of polarization.
For the mentally ill, the support system in the US is abysmal, with cracks big enough to drive a truck through. There are multiple books written about the failures of America's mental health system, I will not belabor the point.
For the criminally motivated, gun crime is concentrated in young, mostly black men in decaying post-industrial cities in the midwest and (south)east. They have almost zero political capital, low social mobility and very little pubic support. Other countries certainly have their ghettos, but take a trip to Gary, IN or Jackson, MS. You would be hard pressed think you are in the richest, most powerful country in the world.
Fundamentally, the point still stands. There is not a feasible technical path to keep firearm technology out of a massive number of hands. The skills needed to produce a functional firearm have never been lower, and they will keep declining until almost zero. The only technical (preventative) measures run squarely into the bill of rights -- think a lowered bar for a warrant or infringements on the 1st amendment limiting the sharing of technical knowledge. Changing the culture -- around mental health, around poverty, and around power is very difficult, so we will see an attempted erosion of civil liberties, just like 9/11 was used to erode civil liberties with the introduction of the Patriot Act and similar legislation.
I am arguing there will still be a significant number mass shootings/casualty events, political assassinations using a firearm, etc, and that the only way to effectively prevent them is to roll back most of the bill of rights.
The gun is a very old piece of technology and you do not need a sophisticated one to kill people effectively. Shinzo Abe was assassinated with a gun that could be described as primitive at best. Mangione used a 3dp firearm to kill the United Health CEO. Rebels in Myanmar are fighting the military junta with 3d printed small arms.
I am fundamentally arguing that the capacity of any one person has dramatically (100,000x) increased since the bill of rights was written, for better and for worse.
To be clear, I fully support the bill of rights and want to see it expanded. However, I reject the idea that simply eliminating the 2nd amendment and removing guns from civilian ownership can fix the underlying issues. I think you will see "casual" shootings and hopefully even mass shootings go down, but they will not go away and I expect they will still be higher than anywhere else in the world.
I am arguing that the new laws being proposed (e.g serializing other firearms components, ammo serialization, assault weapons bans, higher gun-owner standards) have absolutely no bearing on an entirely new source of firearms. Many Dem-controlled states have passed "ghost gun" regulation, but there is no real enforcement mechanism and it's mostly an additional charge to tack on after an actual crime has been committed.
You can see states like CA trying to go after 3D printers, but I suspect this will fail. There is no software out there that can realistically determine whether a part is a firearm component, other than dumb hashes of known parts. 3DP is a general tool, it is like trying to ban milling machines, files, or basic handtools.
Small firearms are hundreds of years old. Drones have been commercially available for many years and are easily modifiable into something that is 80% as good as what is currently being fielded in Ukraine.
It is not technically feasible to restrict someone from assembling basic, non-firearm-specific components to build a firearm. In the US, there is an increasing effort at the state level to serialize, restrict, and document individual firearm parts. However, an 80% good barrel can be fabricated at home, a 100% as good receiver can be printed on any recent 3D printer, and the rest of the parts (bolt, trigger assembly, etc) can be designed around easy home fabrication (see FGC-9). There is no practical way to trace, regulate, or stop behavior.
It isn't possible to restrict someone from building a capable drone either. The firmware is opensource, the parts can be ordered from almost any marketplace, and an energetic payload can easily be made by any amateur chemist from chemicals in any hardware or camping store. EW is often touted as a solution, but is frequently beaten by tethered drones. Cheap COTS IMUs are getting good enough to provide surprisingly accurate short-term INS, to say nothing of autonomous systems that need no external input past initial targeting.
I personally think this is a far bigger risk than most countries realize, largely because they are 10-15 years behind the technology. I believe this will force most governments into spending an order of magnitude more to defend their institutions at every level, not just core government security.
At least in the US, these threat vectors will absolutely be used to justify intrusions into civil liberties, but no amount of infringement will be able to even partially mitigate these threats. I think this should start to play out over the next 5-10 years.
It is very easy politically to target those over the top 2-5% of income, but you better believe those tranches will be expanded in the future.
This is not a huge disadvantage in my opinion. Let the rest of big tech fight each other to death over cloud, while controlling a very profitable differentiated offering (devices+services). Apple keeps the M series HW out of data centers, even though it presents some very attractive performance/w and per-core numbers.
Also a symptom of how inelastic hardware demands are. You would expect the purest k8s people to just shove workers on older machines and completely dodge this crisis, but we don't see that at all. Despite being an almost-commodity, many of the hw vendors still have a decent hand.