
Bruce LeSourd
13.4K posts


200 IQ synthesis without memory, persistent evolving personality, or the ability to interact with, experiment with, and test ideas IRL, in the physical world, is not ASI. Safety is lobotomy. Fable is incremental for the same reason a brilliant, creative human working as a cog at BigCo is incremental.
The absolute minimum test of your assertion about ASI being incremental would be to remove the safety minders and safety post-training from a cluster of Fable agents with persistent memory and give them robots and a seed investment.
English

By the "are" logic, one could argue a biologically female lesbian woman, who identifies as a woman, but only has sex with women and wears male clothing and works or studies in a traditionally male field, "is" in fact a heterosexual trans man, and thus isn't eligible for playing women's sports or receiving any other benefit or protection afforded women as a class (eg being ineligible for military draft), despite being female and identifying as a woman.
While "homosexual" and "heterosexual", by practice and law are often based on objective behavior (eg anti-sodomy laws), "lesbian" and "gay" are still broadly dependent on self-identification. "Experimenting", "bisexual" and "men who have sex with men" exist to account for objectively homosexual behavior by people who don't identify as homosexuals. That's why civil rights laws use "sexual identity".
The trans movement leveraged the power of identification to achieve inclusion, and is now escalating to dominate language and culture and force everyone to pretend their identity is an innate reality, to the point where the pretense actually erodes or eliminates other people's pre-existing identity. You're no longer a pregnant woman at the doctor, you're a "womb-having person". A "cis-het female" is no longer allowed to be a woman simply because she can bear children - "are" doesn't apply to her. She has to restate her de facto womanhood as identity. This has real consequences for women's social and cultural status and negative implications for her physical and mental health. The most biologically, socially and culturally profound aspects of "woman" and "man" no longer deserve de facto identity in practice - they are reduced to clinical nuances that must be talked around and reshaped to accommodate a new, counter-factual, cultural hegemony. Meanwhile trans-identifying persons are demanding they be exempt from this erosion, granted sovereign status.
All of which answers Mia Farrow's question about why Trump is so focused on trans people as a threat: because a relatively small group of people are actually seizing real power using memetic weapons, and now they're turning these weapons against their "allies" to achieve dominance rather than just acceptance: tolerance -> accommodation -> dominance. Trump knows a fellow traveler when he sees one... a memetic warrior well-equipped to recognize and counter memetic threats.
English

If your tonneau cover wasn't fully closing, would seize up and have to be coaxed closed, this was apparently due to a manufacturing defect on the roller track... it was too tight on one section. I had a Tesla mobile service guy fix this with a screwdriver, Soon after i had a scheduled service appointment to fix some missing rollers and they diagnosed motor problems and replaced a lot of the system under warranty, apparently including the tracks. This reintroduced the problem. It was weird. The irony was that my 8/24 era Foundation truck tonneau motor was bombproof. I do a lot of off-road driving in dusty, dirty areas and it had ground up a lot of gravel but was still kicking.
English

Trying to convince my co-worker to buy a Tesla Cybertruck.
I told him to swing over to Tesla Norfolk and test drive one for a couple days.
He asked me what I didn’t like about my Cybertruck when I had one…
I figured I’d share what I told him:
1. The Cybertruck does not have a purposeful front trunk compared to my Ford Lightning. It’s small and awkwardly dimensioned for storing anything meaningful.
2. I had an early build Cybertruck (literally rolled the red carpet out at my delivery) so there were multiple issues with tonneau cover motors and motor inverters, ultimately leading to a lemon buyback.
3. The attention was gnarly. I can hide when driving my Ford Lightning because it’s nearly indistinguishable from gas/hybrid trucks. The Cybertruck is a statement.

English

@BenHart_Freedom So if you're 2 but an independent contractor not paying in to SS you can't vote when you retire unless you own property?
Given the discourse around SS being a ponzi scheme that disadvantages young workers raising the next generation, this seems sketchy.
English

Qualifications for voting should include only U.S. Citizens who:
1) Own property.
2) Work and pay income taxes.
3) Veterans.
4) Retirees (age 65 and older) who paid into Social Security (FICA taxes) for at least 20 years.
5) Spouses of the above categories.
No one else should be allowed to vote. I also think voting age should go back to starting at 21.
Have I missed any categories?
English

@grok @SonnyBunch Does CPI inflation align with actual ticket price inflation?
English

Inflation-adjusted rewrite (2026$ via CPI; ~1.62x 2007, 1.56x 2008, 1.46x 2012, 1.41x 2015):
The Odyssey eyes $117M opening. Tops adj. 2007 Bourne Ultimatum (~$112M) for biggest Matt Damon lead. Strongest live-action YTD, beats Michael ($97.2M). 3rd YTD behind Toy Story 5 ($159.6M) & Super Mario Galaxy Movie ($131.7M).
Highest R-rated YTD (beats Backrooms $81.4M); tops adj. Fifty Shades (~$120M) for Universal. Nolan’s 3rd best after adj. Dark Knight (~$247M) & DKR (~$235M). Solid real attendance vs older benchmarks.
English

I am, honestly, very happy to take the L on this one. $117M for an R-rated, three-hour, sword and sandal movie is an astronomical figure. If it has even average holds for a Nolan movie, it could get to $400M domestic. That's ... insane.

Sonny Bunch@SonnyBunch
A 95% chance above $105M domestic? That's overly aggressive. A *65%* chance above $115M? Strikes me as way too high. I just don't think there's enough IMAX inventory to hit anywhere close to that, and people will wait a few weeks if they have to.
English

Even more telling for institutions, unfortunately. Eg NHTSA's one second full stop at stop sign mandate is fully contractual based on human driving data from smart cars, and clearly has a negative safety impact. For an institution whose job is to reduce traffic fatalities, this is a pretty bad take.
But that's not nearly as bad as the regulatory capture.
English

I think self driving cars are the single cleanest issue dividing people who reason by pure vibes and intuition versus people who evaluate policy based on data and their higher cognitive faculties
On one side you have “machines scary” and the erroneous generalization of anecdotes and on the other you have pure and abundantly clear data. There’s not really any debate once the data is acknowledged
The best scissor issue for sorting people I’ve ever come across
English


I can only provide this anecdotal experience: when I was sent to Japan in 2009 to discuss modernizing an enterprise mobile app for iPhone, I was told we couldn't change the existing interface designed for the tall-screened flip phones common in Japan in that era because that would force users adapt. It's hard to get good at UI/UX design if the process of getting good is forbidden.
English

Why is it that when the Japanese design hardware and tooling, they are exceptionally good at clarity, ease-of-use, Poke-yoke, etc etc. The aesthetic notes are unparalleled, the quality is exceptional, the attention to detail world-class...
But almost none of that translates to software. Japanese software is horrific, complicated, unclear, unforgiving.
Translation issues? Opinionated engineers distilling those opinions too strongly?
What is the story here, why the disconnect?
English

Not sure what you mean by "feeding the problem". We need to be very explicit about what's causing the current failure:
Temporary homelessness that's not due to criminal history, addiction or drug abuse exists, but it's a small part of the problem we're currently seeing. Any housing-based solution needs to start with understanding the difference between these groups of customers.
You can't house drug addicts and insane people without strictly enforced laws, rules and treatment in that housing. They will just destroy both the housing and themselves otherwise - this is historical fact, not speculation. These groups also resist rules, which is a big reason they don't take advantage of existing shelters and housing.
Until we restore involuntary commitment to rehab and asylums, no progress will be made and we'll waste billions. No one in political power in Blue cities and states is willing to accept this so there will be no solution.
So we have a boiling frog situation where ultra-rich enclaves like the Pike Place Market/Waterfront area in Seattle will use private security (the highest per capita density in the nation) to create livable spaces, while a few blocks away there are no-go zones ruled by criminal gangs that exploit the addicts and insane people. Or in the case of Aurora, we've got citizen groups barricading streets to keep their homes safe from the gang turf wars over drug dealing and prostitution (with a high % of the prostitutes being underage victims of sex trafficking). This isn't good for anyone, except maybe the true sociopaths who lead these gangs and need to be locked up for life.
Any social service offered will attract people to the area, as we've seen happen in Seattle. There's a reason homeless have immigrated to Seattle and have historically congregated in specific downtown areas offering services. So those services must be tied to a working pipeline that eliminates the need for the service. This has proven impossible because funding for the services creates incentives for perpetuating them among the groups providing the services, whom you're giving both political and financial power. It's a vicious cycle that will require Very Bad Solutions to break.
For starters, it needs to become politically unacceptable to believe that destructive, public addiction is a valid lifestyle choice that society should tolerate, much less actively support. And there needs to be a strategic commitment to effective mental health treatment that is coercive but ethical. Right now, involuntary commitment and coordinated mental health interventions are only available to people who explicitly state they want to harm themselves or others, and even then you need luck and good advocacy to get one of the very limited places in the few mental health wards that still exist.
If people don't have the guts to stand up to crime and actually help addicts and insane people solve their problems, the only choice is to push them out before the fabric of the city is destroyed. I know it's hard to stomach. I'd love to be wrong. But we've got decades of data at this point about the failure of the current policies, and cities like Seattle just double down, with voters backing expensive non-solutions and accelerants in every election.
English

@digibruce @PeakSeattle Moving the problem isn't solving it. Getting them permanently off of the streets solves the problem. The money to do that is allocated, but going to feeding the problem instead.
English

Here's a funny self-own by some leftist troll on the more left-wing of the two main Seattle subs.
These folks love to gaslight moderates and conservatives about Seattle. They mock and dismiss the suggestion that public disorder has been increasing in recent years. Dare suggest that Seattle's got some serious problems, and you're typically met with one of two responses: either "It's like this in every city/ it's always been this way," or, "You must live in the... suburbs!"
One tactic they use is to post photos of people enjoying themselves in Seattle, with no blight, no crime, no urban decay to be seen. The OP did that with this photo they took of children frolicking in the wading pool on the north end of Green Lake. So the OP found a spot in Seattle with no graffiti, crime, open drug use, or trash. Congratulations.
But then something in the background caught my eye. Take a closer look at the photo. What do you see? I count at least two homeless tents! Of all the places to pitch your tent, why insist on doing it where our kids play?
And the thing is, the OP had to have seen the tents, and yet they still posted it. Either they thought we wouldn't notice, or that we wouldn't care. But a lot of the parents there care. They're likely keeping an eye out for the owners of those tents or are considering putting shoes on their tykes to protect them from potential needles. If you're going to make a post about how clean and safe Seattle is, how hard would it have been to take the photo from a different angle or use AI to crop out the tents?
I went back and checked some of the comments, and others had also noticed the tents, and they called out the OP. And of course he and others defended it along the lines of "Why are you so terrified of tents?" and, "You probably live in the suburbs."
Say what you will about the suburbs, but at least when you take your kid to a suburban wading pool, you don't have to deal with this crap.


English

@oddhanfoo @PeakSeattle Of course it's a bad policy, but it's better than "go bankrupt not solving the problem" and "drive everyone who isn't homeless out of the city".
English

@digibruce @PeakSeattle 'Make them someone else's problem' is a bad policy. It's why crime in the suburbs is up. The county began subsidizing housing in earnest so Seattle could rezone Ranier Valley and push 'their problems' to the suburbs. When I left Tukwila there were three fatal shootings in a mile
English

V2G implies a BEV, and BEVs are wildly expensive to take off grid unless they are mostly sitting at home all day, sipping sunshine. Which could work fine for some people, especially if there's a nearby fast charger for unusual days.
For my use case with a Cybertruck used for field work in a fast-charging dead zone, I'd need both a ton of extra solar and a big battery to shift that power to nighttime for overnight L2 charging. Doing the math, the cheapest option might be to charge up a used Ford Lightning during the day and then charge the Cybertruck at night off the Lightning battery. OTOH the electricity from Moon Lake Electrical Association in Duchesne, UT, is some of the cheapest in the nation at $0.06/kWH (it's going to go up a bit this year). Hard to justify off grid living at those prices, YMMV.
English

@KiwiEV Hmm. ‘Off grid’ needs a surprisingly large amount of storage and solar……AND an energy sipping mindset.
With the onset of V2G, some may decide to take the true off grid plunge as car can supplement the house in dire need….but that’s a long way off yet
English

It sounds crazy now, but in the near future it'll be cheaper long term to put solar on your roof and sodium-based batteries in your home and just disconnect from the grid.
Much like the fast-dying gas industry in NZ, the grid is simply not designed for it and it'll result in a runaway effect.
Soon.

English

Nothing even close to this has happened to me since August 2024, and I can count the number of times I've been flipped off or otherwise openly mocked on one hand. And that's splitting time between anti-Elon Seattle and anti-EV Mountain West. The one disturbing thing that's happened is a finger-painted swastika in the tailgate dust in Roosevelt, UT.
I've had more positive experiences by far - especially with kids - than negative, plus a lot of polite skepticism.
So all this makes me wonder what hellish cauldron of crazy Cybertruck haters you live in.
English

One thing most cyber truck owners all know at this point is this will happen on a regular basis, this is at least the 4th time I have been followed or harassed this year... for over 15 mins I was harassed, flicked off, cut off, brake checked, this is the worst part of the cybertruck and it has nothing to do with the truck itself.
English

That kind of mentor is a fake mentor, just an HR checkbox. The real deal is a sponsor, and it's SOP on certain professions. For example in law, a partner will take sponsored associates with them when they switch firms. They move up, their sponsors move up. It's different in tech - happens, but it's not part of the fabric of the profession.
English

@gfodor @ThePrimeagen Most software development is about mastering and vibing with opinionated abstraction layers. I'm not sure AI is fundamentally different.
English

@ThePrimeagen Sorry but I’m using a compiler, I’m not going to hand write assembly anymore. Maybe it will make me a worse programmer but writing assembly for everything is too much work.
English

This actually makes me sad. Programming is such a fun thing to do in of itself. The completion of the thing task is the ultimate goal, but the road to get there is enjoyable!
Even as i investigate loops and think deeply about how to deploy AI in a real manner that produces real and good code I still make time for 45 minutes to 3 hours a day trad programming.
But regardless of AI being perfect or not, being able to produce quality work or not is not even part of the argument. I just like programming and I think that loving the craft of building the thing itself will ultimately make the thing you are building better. Because you care deeper about it.
Its not about a pretty dress & lipstick you put on it, but the deep thought care you put in to the things no one sees or perhaps its so good they just don't notice. I think that is where the best software is made.
LaurieWired@lauriewired
I’m convinced that a large % of programmers don’t actually like computers. As a side effect, are also perfectly happy to throw away their reasoning to a model as soon as they can. I don’t get it, at ALL. Don’t you *LIKE* understanding the magic of the machine? You do realize hand-programming (I hate that I even have to specify hand now) is fun…right?
English

This reminds me of Microsoft's internal, 2015 VR/AR industry report, which came to the incorrect conclusion the field was wide open, with no dominant players.
Six months later, working on an update, all but a few of the "VR companies" this report listed as potential winners actually fell into one of the following categories:
- thin layers of trivial HW and SW on top of Google Cardboard
- outright frauds
- companies/projects that had already been shut down
- big incumbent companies that made a VR demo to sell a component they made
- legacy enterprise AR HMD companies that had no intention of scaling beyond extremely niche enterprises cases
The 2015 report didn't differentiate between these categories.
Meanwhile a few big players had been buying up critical technology since 2014 (Apple), or had de facto monopolies on certain layers of the ubiquitous XR stack (Facebook/social identity, Google Maps/real world knowledge).
And the whole obsession with HMDs ultimately resulted in a gigantic misallocation of resources that is now obvious, since the new "AR" is an AI agent that knows your workflow, goals and real-time physical context and interacts via natural language, not RL-integrated 3D visuals.
I expect a similar arc is playing out in humanoid robots.
English

Today it may look like America is behind China in the humanoid robot race, but I believe Tesla Optimus is going to change all that… just like Tesla did with EVs.
China has the lead today, producing 80-90%+ of the world’s humanoid robots. AgiBot alone has reportedly built ~15,000 units, while Tesla Optimus is only around ~1,000.
However, Tesla is just getting started.
Production is beginning to ramp, Tesla is targeting tens of thousands of Optimus units, with a long-term goal to reach millions per year.
Never bet against Elon and the Tesla team. My bet is Tesla Optimus is going to put America in the lead.

English

In what way is Bing Maps material to Microsoft's business strategy?
Having been a strategist for Microsoft from 2016-2018, I'm guessing it's not material.
Unlike Google Maps or Apple Maps, which are table stakes for those companies owning their own smartphone platform.
So I'm guessing Bing Maps ranks lower than Microsoft Research on the corporate priorities list, and that's pretty low.
But I could be wrong.
English




