pixelready

This works the same way irl too. The US is one of the most heavily propagandized populations in history. You have to be talking to a very progressive, very politically savvy group to not constantly run into the usual knee jerk canned responses the second you bring up any criticism of military spending, foreign policy, or the panopticon. Probing of those responses in family or social settings for actual justification shows there’s no first principle reasoning behind them, just a vague unease about discussing the topics, like they’re sacrilege.

I’m sure arrogance is a non-zero factor, but decades of anti-labor propaganda is likely a bigger contributor.

Ask 10 people in the street about their opinion of unions, and 8 of them will regurgitate some vague concern about union leadership corruption, protection of “lazy workers” or exorbitant dues, rather than cite one of the many hard-won historical victories against exploitation or unsafe working conditions over the years. That’s no accident.

Assuming high-quality models are around for the foreseeable future, and at this point I’m convinced that they will be. I think we’re moving into an era where breadth and hybridization of skill sets will become more valuable than pure technical depth. We’ll need more “wide T” shaped people that can exercise good UX, product, and architectural judgement and fewer “long T” shaped technical purists.

This puts large orgs at a disadvantage because they are built to coordinate work across a collection of narrowly scoped roles and keep people in their lane, so they aren’t built to hire for and leverage broad skillsets.

It’s requiring all my willpower not to take out a second mortgage and build a gallery wall filled with these. That image transition is gross but I guess it takes a lot more steps to move all those pigment layers around.

Grifting, purchasing loyalty, and building a police state panopticon in case of inconvenient election results. A real dystopic trifecta!

Keep people in “system 1” thinking to control them. Given the palpable shift in online conversation since a couple of Russian datacenters got blown up, it’s clear that although these fault lines always existed in US society, fringe opinions have been massively signal-boosted by bots for quite some time. I suspect there’s been a psychological Cold War going on online for at least the last decade or two, with a whole ecosystem of grifters and ragebait view farmers bottom-feeding off of state sponsored PsyOps, leading a corrupt inner circle of elite opportunists to take advantage of the resulting anger for personal gain. We’ll probably never get to tease apart what was a natural side effect of social media and what was engineered deliberately but I bet the story would be every bit as fascinating as the Manhattan project.

We just need to rush Artificial Personality Matrix research so we can complete our Synthetic Ascension. Habitability will be a non-issue.

I think we want algorithmic feeds, we just want humane ones instead of adversarial ones. Give me a feed that prioritizes close friends and family first, important milestones over trivialities, impactful news over ragebait. Give me a highly curated version of Reddit with HN quality moderation. Most people would be very happy with that, but it runs counter to the advertising incentives and engagement metrics which drive the industry.

Hard agree. Modern frontier models will give you a surprisingly solid first pass at a concept if prompted with a clear initial description and a handful of well-guided feature refinement rounds in an afternoon. That said, if you have a high bar for things like concise, well-architected codebases and strong, intuitive UX (the two main components of “polish” imo), you’ll find it takes a long time to get from Pareto-decent to truly polished.

I get where you’re coming from, but I think if the progression of technology has shown us anything, it’s that there will be constant competitive and social pressures being applied to use these tools more and more and that pressure will race far ahead of any privacy, consumer protections, public education and societal wellbeing counter forces. So opt-out ends up being a rapidly shrinking iceberg in practice.

It feels icky, and my default is not to side with megacorps engaged in blanket surveillance, but I can’t really fault Anthropic here. The correct setup should be a law that protects this sort of interaction with a chat bot as privileged, along the lines of HIPAA-mode. Use a classifier or some sort of “private mode” toggle to tell the platform you’re engaging in privileged communication ala dear diary, and then a much higher legal standard needs to apply to protecting that data (no training, same protections as doctor / psychologist interactions). Even a therapist has a legal duty-to-report in certain situations.

I tend to think this way as well, though I find team stochastic parrot is shrinking by the day. It is very difficult to explain it to people who don’t have at least some grasp of what’s going on inside an LLM, and how they’re trained. We have no psychological “immune system” for something that mirrors our ability to deliver an apparently well-reasoned response to a question in our own language. All our instincts tell us to assign it sentience and treat it as an independent actor with it’s own internal motivations and personality. Even I catch myself doubting that a pile of matrix multiplication isn’t “alive” in some way after using it for a while.

Some are conditioned, but I think there will always be a subset of people that just feel like they are the “right people” that everyone else should emulate. The system feels like it’s there to protect them, strong authority makes them feel safe, and they have a strong instinct to conform. They’ve never been on the wrong end of a political witch hunt, colonized, or felt like they were being attacked for wrongthink, so they don’t understand why everyone else can’t just be “good, law-abiding citizens” like them. Seems utterly alien to me, but I’ve known more than a few.

If these cloud providers had needed a standard customer acquisition strategy to grow to their current size, hard caps and other “training wheels” features would already be in place to get people interested in and comfortable using the platform, with the hope of eventually getting a foothold into Enterprise like most SaaS startups have to do (“enjoy our product on a side project and then recommend us to your CTO!”). But AWS and GCP got to start as in-house providers for their own constellations of massive sites and back out from that to serving other hyper scale businesses first. The lack of friendly on-ramps and starter account features is a reflection of that origin more than anything.