Twitter stopped publishing stats on its user counts when it went private, so we don’t know if those have gone up, down, or sideways. We do have some numbers for Europe since about 2023 because of reporting mandates, and they seem to show a downward trend.
We could split hairs about numbers or acknowledge that X (and TikTok, and Instagram) still owns the zeitgeist. No normie gives a shit about BlueSky and Threads is anemic considering they gave every Instagram user a free account. I'm being real here.
Certainly it didn't implode and is not failing but it's not clear to me that its user base is increasing from pre-Musk. And Threads supposedly has a similar number of active users.
Twitter had around 230 million monetizable daily active users in 2022 before Elon Musk acquisition. They didn't publish monthly active users. [1]
Twitter IPO S-1 in 2013 said 500 million tweets per day from 215 million MAU and 100 million DAU. "We currently estimate that false or spam accounts represent less than 5% of our MAUs." [2]
2024 Elon Musk claims 600 million monthly active users [3]
2025 SpaceX S-1 Filing 520 million monthly active users. [4]
2026 Also SpaceX S-1 Filing 550 million monthly active users. 350 million daily posts. "The total number of supported accounts may include fake, spam or bot accounts if they are active" [4]
Threads hit 500 million monthly active users in June 2026 [5]
The oil that lubricates the combustion chambers and pistons gets burnt and filled with nasty combustion goop. An impressive amount of cleverness goes to minimizing this, but the oil that sees the combustion chambers still goes bad at many times the rate of any other lubricant.
Even on an ICE vehicle, you have transmission oil and differential oil and lubricants in your air conditioning compressor and brake hydraulics and grease smeared on CV joints and door mechanisms and pretty much everything else that moves. But you still talk about "changing THE oil" and mean the oil-that-sees-combustion specifically (along with, typically, the filter that fights the constant battle of getting the worst combustion products out of it). I personally think that makes it fair to say "EVs don't require oil changes" because even though they don't magically eliminate the other lubrication systems or the need to maintain them, they do eliminate the one that is so much more obnoxious than the rest that we commonly called it "THE oil change" before this was ever a debate.
It does, plenty of things inside the EV require all kinds of oils, just the maintenance cycle isn't gallons and gallons every x months. You are looking at years.
I got service centers trying to rotate my tires every 3k miles because "ev". The oil change as well, oh you need to change oil every 5k because... because why?
You got brake fluid, calipers, but it’s not even close to a regular car.
The dealerships and shops are doing every one a huge disservice by lying about how frequently engine oil needs to be changed. It’s in the manual but even that should be on page one. And then they set reminders on the car inside whatever last changed or when near, miles. So they’ll set it for 5k instead of 6-8k as the manual says and reminder at when you hit 3k. Like cmon.
Or tell me that I need a balance every 3k miles in a Tesla. Shameless.
Engine oil is pretty much the only critical one because of the tolerances between the piston and the cylinder. Loose tolerances allow bypass (inefficiency), but tight tolerances require good lubrication (friction). Engine oil is also unique because it’s exposed to oxygen and combustion byproducts.
The other components just aren't as sensitive or exposed to as much abuse. Often, really only needing replacement if they’ve leaked out.
Not op, but brake fluid comes to mind, power steering fluid, transfer/reduction box. All depends on which ev you are talking about. Things like cv joints, axels and other rotating parts also require greases, again depending on make and model. Not that those need yearly replacement but they do need maintenance.
Brake fluid isn’t oil (generally speaking). EVs have electrical motor-driven power steering, not hydraulic. You scored correctly on the reduction box, which will need changing well after most people sell the car (>100K miles).
I find it hard to believe that someone somewhere isn't making an EV with a electric pump to supply a traditional hydraulic steering system. If I had to guess it would be Europe because they love electric over hydraulic steering.
“I haven’t brushed my teeth, but I don’t have a cavity - so I’m good”.
Don’t get me wrong, my “fun” car (an old, beat up Jeep) is going on 2 years with the same oil (I barely drive it), but I known I need to change it soon.
Oil degrade with oxygen. Every time you drive that car, the oil are being re-oxygenated and continuing to degrade (along with being exposed to combustion byproducts).
Also, the type of wear you get often isn’t noticeable until it becomes catastrophic. In my case, a 30 year old car that’s already on its last legs “catastrophic” is more like “time to finally drag it to the scrap yard”.
Motor oil is a wear product with a lifespan. It will become saturated with microscopic particles of metal and end up with lower lubricity. You are just slowly grinding away all the metal off the moving parts. Eventually the tolerances will fail, usually a rod bearing, leading to self-destruction of the motor.
You can find these stories in the mechanics subreddits, typically it's some Toyota came into the shop dying at around 60k miles with the factory oil turned to sludge.
Oil is a consumable with both a use lifespan and a time lifespan. So it needs to be changed at the lesser of either of those intervals. Same with tires, which a lot of people don't realize.
They do have some gearbox oil that is usually considered to last a lifetime but some sources advise changing it every 100k miles or so. This is analogous to transmission fluid in an ICEV.
The seal between the piston and the cylinder isn't perfect, so when the engine runs combustion products will mix with the oil, making the oil eventually unusable. That's why ICE engines meed oil changes so often.
And in fact it is necessary for it to be imperfect, and the oil is what lubricates the interface. So it is necessarily exposed to lots of heat and pressure.
There is oil for the gearbox, but there's no scheduled maintenance for it at all, and in practice, people regularly get hundreds of thousands of miles without ever needing to think about it.
On a DSG gearbox the interval is 65k km. You should also change the transmission fluid in a regular automatic and a manual transmission. Some cars have a service interval for those transmissions and some don't. The "lifetime" oil statement is a lie.
Bearings and shocks have grease, sure. But those things last decades after installation. They're probably two orders of magnitude lower in volume consumption than engine oil, which gets contaminated and disposed of at the rate of like a gallon or two per vehicle per month.
Like, the price differential is so high that the high end artisinal lubricant market is dominated by chemically synthesized products that aren't strictly petroleum at all (though many of the reagents used to manufacture them probably come out of the petrochemical world).
There are plenty of things that still need oil and grease in EVs still, but it is in such low quantities that we could synthesize most of that from air and it still not being a cost of any concern.
Correct. ~140k miles on my 2018 Model S, tires, wipers and wiper fluid, and one battery pack heater replaced (deprecated, no longer used in newer builds) are all that have been done. Motors are lubricated at manufacture and sealed for life.
I’m still on the original brake discs and rotors luckily but will likely need to replace them soon from corrosion and warping versus consumption. Which is very cool imho.
Transmission/differential oil tends to last a very long time (150-400k, unit doesn’t matter). Automatics require relatively frequent oil changes in comparison, but they’re using the oil to transmit power.
Lots of greased parts, but again, the grease will last about as long as the part (mostly because the grease falling apart makes the part fail), which usually lasts the life of the car.
I would imagine if there is oil in the electric motors that is “lifetime” but test really probably 300k miles, so if you drive the car into the ground, you probably would want to service that at some point, assuming they are serviceable without destruction.
The only significant oil is for the reduction gearbox, but as it doesn't change gears the wear is very low. First change will pull out some metal I'd guess, quality of machining dependent. Every 60000km/2Y, 1-2 litres.
Batteries may be immersed in a high dielectric thermal fluid, which will experience some breakdown but over a long time period, more than 10Y.
You’ve got parts with oil that have to be done every 10 years or are completely sealed into the part and are expected to last as long as the part does.
And idea like this is why EVs will still have poor depreciated values even if the battery holds up. Yes it is common for people to ignore all the non-powertrain maintenance items on cars before EVs, but at least in ICE half of them regularly dripped oil down its body and lines and still semi-regularly had somebody kicking the tires and shit when the oil was changed to try and sell you a repair before it fails.
Body, wheels, and suspension components all require maintenance. Just because those components have a less predictable maintenance schedule doesn't mean they don't crap out if problems aren't identified and taken care of. And unless you are driving a short range smart car, EVs weight more than most ICEs, which accelerates the wear and shortens the lifespan of many components.
Don't get me wrong, EVs should be more reliable than ICE still, but in my experiences motor and engine problems are rarely the death knell of a vehicle, body damage and rust are. There are always enough people crashing or junking their cars so that even less reliable motors can be easily had on the used market for cheap, if you don't just swap to something better to start with, but once you get significant body rust 95% of mechanics won't/can't deal with it and repairing metal body shell is a level of skill and tooling that most DIYs aren't going to even attempt.
I don’t understand this criticism. I have owned an EV for five years now. The app and the vehicle display clearly let you know when you are due for scheduled maintenance.
Due to how online the whole platform is, it’s so much better than my older ICE car that has a cryptic wrench light on the dash.
Did I mention that the intervals are also so much cheaper? Tires are the one standout but pretty much every thing else is a dream
There are multiple people in this thread talking about about how their EV requires zero maintenance despite owning them for years, which is just the admitting they are setting up a ticking timebomb of maintenance tasks that are being neglected. And then years down the line they will wonder why their model car has depreciated so poorly despite their battery being in good condition. And they will likely blame ignorance of EVs, rather than considering it is the average owners negligence of maintenance tasks on that car.
What ticking time bombs? Please be specific. I'm an EV owner who has done no maintenance in 26k miles. I want to know which system, if I don't do maintenance, will create some hidden wear which will progress silently until too late.
Shocks, springs, tie rods, steering systems, motor mounts and any other rubber parts, cooling systems, braking system, wheel and possibly motor bearings.
Nothing special for being a car, but an ICE and transmission weren't the only things on cars that move and wear.
There's no maintenance to shocks, springs, tie rods, motor mounts, or wheel bearings. Every one of those announces their failure with clicks or clunks or bouncy ride. I'm not sure when chassis lubrication went away as a maintenance activity but I'm guessing around the '80s. I can remember using a grease gun on my parents' cars but none I owned. EVs obviously eliminate all the engine internals, so no spark plug replacements, no ignition modules, no accessory belt, no timing chain or belt, no engine air filter, no oil filter, no manifolds, catalytic converters or mufflers.
Brakes need to be inspected but my last ICE vehicle went about 90k miles between needing pads. My EV has sealed drums in back (a choice VW made to reduce friction when brakes aren't in use and I think trying to slow corrosion of the components) and conventional large discs in front. If we have issues with the front brakes it will be due to corrosion as I don't expect to put enough wear on them to ever wear through the pads. In normal driving the brakes are only applied at full stops.
I don't expect EVs to have motor mounts like an ICE vehicle, as the motor is a few orders of magnitude smoother (commutation ripple vs cylinder firings) and shouldn't need fluid-filled rubber isolation mounts.
If I think of what I've had to replace on my cars over the years, half-shafts and CV joints and wheel bearings might now be the least reliable mechanical things on the car, plus the oddball failures, like a clock spring in the steering wheel or a window motor.
So I still have to keep tires inflated, rotated and balanced, keep good wipers installed, change the cabin air filter occasionally, and listen to the rest.
My source for my opinion is having done the vast majority of the maintenance and repair of our cars over roughly 4 decades and more than a million miles.
Not really. Most stuff is sealed for life. So it's more replacing whole parts when they fail than changing the oil in something. But many bearings will last hundreds of thousands of miles and never need changing for the entire life of the car. Quite amazing really.
Only if you interpret statements as being binary logic.
"seem to be" carries semantic meaning here: I'm stating my interpretation of the situation based on data we have available (which is limited) and my prior.
Put another way: "We can't say that for sure, but my money is on it not being a simple case of intellectual property theft"
Imagine if they broke it down to each distinct source, that'd be several billion cases of copyright infringement (though it's going to be determined by what courts think and that often comes down to "who can afford the best lawyers" in practice if not intent).
Apparently if I use lib-gen, that's copyright infringement and I'm exposed to legal risk but it seems fine to download all of it if your intent is "train an AI" so far.
Other people are allowed to have their priors, too. Even under a non-informative prior, the weight of the evidence (Tristan's account, OpenAI's announcement, and Bubeck's "denial", if you want to call it that, plus multiple other mathematicians coming forward with similar experiences) pushes the probability mass toward some degree of impropriety.
What exact evidence are you incorporating into your prior to come out with this posterior?
We are giving you an opportunity to correct yourself. You are instead trying to make your nonsensical statement make sense. Not only does the first part of your sentence literally contradict the second part:
> We don't have enough accurate knowledge to say [one way or the other], and it doesn't seem to be the case at all [based on our incomplete knowledge].
But it is in no way equivalent to this:
> We can't say that for sure, but my money is on it not being a simple case of intellectual property theft
Based on: my experience working on AI for 32 years, including a decade at Google including working on large-scale model training systems that used user data and complied with various user policies around data retention, along with a few decades working in science/tech making decisions around ambiguous data.
In short, I have a well-tuned intuition and a huge set of priors, and applied them to the limited knowledge we have about this situation.
The human mathematicians didn't solve the Navier-Stokes problem, they solved the Euler problem. And they were extensively using LLMs to drive the work, as described in the Buckmaster statement.
Any way you cut it, this is a major achievement for AI, besotted with human drama over whose prompt should be recognized by the history books.
We understand the proof (most math from all of history) -> We understand how the proof was made (computer-assisted proofs like the four-color theorem) -> We have to trust the computer's explanation for how the proof was made (some LLM proofs)
Is that really a problem? The only risk in a Lean proof is the LLM finding a bug in multiple kernels (2 with the standard setup) that would let it add a new theorem without legitimately proving it. If you have reasoning traces or send a second model to look at the proof you can easily detect that and file a bug report.
For the rest you only need to check that the formal problem statement matches the actual problem and the proof is not using extra axioms. That's easy to check manually or with a simple script.
Many mathematicians care about understanding the method and the tools used more than the result. LLMs have a large advantage in memory and reading speed, you can feed them the entire internet and published literature at training time, and then make them check dozens of papers in a few minutes at runtime. They might not be very creative, but they've seen that obscure 1990's theorem in the training data that happened to use an idea that might just work again.
Reading the LLM's 12 MB lean output is a bit like trying to maintain a codebase that has been fully vibe-coded as a single giant file, by hand. Even if it's possible in principle, no one is going to enjoy that.
Qwen is very much censored. Just try asking it about Tiananmen or how to build a bomb. But it is nice that you can experiment with it locally without having to worry about your account getting nuked
You are misunderstanding what they said, they are saying you can use uncensored variants of models like Qwen when running locally. There are quite a lot of people working to "uncensor" open weights releases. It seems to work although it would be nice if some third party was benchmarking the uncensored variants regularly to give us an idea of how well retained their skills are.
What is terrible about charging at home from a regular outlet? As long as you average <50-60 miles a day, like most people, you basically never have to go to fast chargers (or gas stations). It’s amazing.
I have a standard plug (not even three phase) and charge 30 kWh in ten hours (overnight). That's around 90 miles, which is plenty.
If I'm going on a long trip, I just make sure I start charging from > 50% the previous day, and I'm good. It's trivial to manage the state of charge when you have a charger at home and a planned trip coming up.
In the U.S. our standard outlets are 120V, so you only get about 1kW or so out of them. Plenty for most people, but annoying if you need >50 miles of range a day, or if you want to boil water in an electric kettle. The situation is much better in Europe, and most of the rest of the world, where you have 220+V
I’m convinced super-fast charging will be a game changer, because it will eliminate the ~only remaining downside to EVs. If Tesla had charging tech like BYD the market in the U.S. would look very different
I don't buy it. "Recharging == Fuel" is the wrong way to think about it.
Think about your phone. Do you plan how you'll charge that quickly? Do you know where you can get a quick charge at a local mall while ordering coffee? Nope, it just gets charged over night when you're asleep, no problem.
I agree charging at home/work is best, and ideally we would have ubiquitous slow charging at apartment lots, workplaces, and on the street. But that’s not really realistic right now in the U.S., and more importantly, it isn’t under Tesla’s control. Whereas they could have solved the problem for people with less charging access by making their fast charging faster like BYD has done.
With respect to your phone analogy, if I weren’t able to charge my phone at home or at work, I would definitely be interested in being able to fast-charge it at some conveniently located business
I think you're both right. An EV is wonderful if you can charge at home, and a chore if you can't. Fast chargers help, but they're expensive and still not as fast as gas.
If you have a wall plug at home, an EV is basically always full. I'm shopping for a flat and one of my non-negotiable requirements is a parking spot with a plug. The one I'm looking at now even has a three-phase plug, which is a nice extra, but not really necessary.
Yes I do get the fastest charging speed and highest battery capacity phone, which are Chinese now too like OnePlus. Once you know you can depend on your phone for a few days, it's a game changer.
reply