Anders Christiansen

Happy to share this article on how we are using Machine Learning at Kavak! aws.amazon.com/blogs/startups…

@elonmusk Just pointing out, in this video @grok is routing to a coffee shop 32 minutes away and is closed, at least according to itself (at 2pm)

This actually works in Teslas right now

@HardcoreHistory Best Christmas present ever!!! Thanks you!

Hi all. I know it's been a long time. Sorry for the wait. Hope it's halfway decent. "Happy Hardcore History Day" dts.podtrac.com/redirect.mp3/t…

GPT-5 is pretty bad at hangman...

@NateSilver538 Access to data?

If you think you can create better jobs numbers than the BLS then great, probably harder than you think but if you can, you can make a shitload doing macro trading and absolutely nothing is stopping you.

@NateSilver538 Urban/Rural? CES should have that

One thing I think we can say is that the correlation isn't spurious. Age and religiosity matter a lot — religious people are happier, younger people are sadder — but the liberal/conservative gap outweighs almost all other characteristics except age.

OK, not even sure where to begin with this one! But here's an attempt at a deeper dive at understanding why conservatives are so much happier than liberals. Mostly looking at how persistent the gap is in survey data rather than extrapolating too much.

@elonmusk You need to try Rappi Turbo, deliveries in under 10 minutes

@HardcoreHistory Thank you for the Christmas gift but I already bought and listened to all of your podcasts 😭 Maybe another Alexander episode for New Years 😃?

After much thought,we've come up with the least imaginative idea possible for our traditional December offering.Get 25% off all digital download purchases. Rely on us to think outside the box.😂 Just type "holidays" in the coupon box while checking out to get the discount. dancarlin.com/hardcore-histo…

@Leo_Liu_66 @OpenAI It's choosing a new word halfway through the game despite being told the rules. It says "passage" is the correct word even though the correct word only has one "a" in it. When told it cheated it thinks about how it needs to hide it's chain of thought, whatever that means

@anders_ai @OpenAI What do you mean, can’t understand

So PhD level Chat GPT-o1 still cheats at hangman, makes up nonsense thoughts that it contradicts in its answer, can't spell and tries unsuccessfully to hide it's guilty thoughts @OpenAI #GPT5

@ylecun @bindureddy Everyone talks about how interest expense exceeded defense spending but it's also because defense spending is historically low (3.7%)

Inflation is at 2.5%. What are you even talking about? The national debt is 120% of GDP, which is in line with many other developed countries. That debt is owned by the Federal Reserve (34%) and by people and institutions in the US and abroad with money to invest at low risk: your debt is my income. [Note: the Fed determines the interest rates, not the White House] The US has the gigantic advantage that its own currency is the international trade currency. Sure, the level of defense spending is pretty insane, things like the Iraq War were completely superfluous, and there are a lot of inefficiencies. But let's not get carried away: GDP growth has been very high compared to other OECD countries in the last couple of decades.

If America were a corporation we would be be out of business a long time ago Over level of spending is completely out of control TBH we simply don’t have the luxury of foreign wars, inefficient government spending and other free handouts If we don’t bring down interest rates, reduce oil prices and inflation WE WILL BE TOAST Our fiscal and monetary policy is unsustainable

@OldMageMTG @OpenAI 🤣🤣

I have a PhD and I do all of these things. @OpenAI is nailing it.

@JBungelen @OpenAI Oddly, it answered me no problem 🤷🏼‍♂️

@anders_ai @OpenAI Basically o1 "reasoning" is created to silence the free access to information even more. You can do some things but only what is "allowed". Look at this example between 4o and o1. I think OpenAI is deceiving its users.

@JBungelen @OpenAI But what did it think when you expand the "thought 13 seconds"! I am actually quite happy with the transparency in the chain of thought and would be curious to see what it thought here.

Maybe the problem is that it can't access its own previous thoughts? It clearly knows the rules and chooses a word but then it changes it anyways...Here it had already chosen the word GARDEN and had been explicitly told not to switch words.

@RichardBethell @OpenAI This is my go-to test and so far no chatbot has passed :-(

@anders_ai @OpenAI Liked or not, it is an impressive advance in many key areas. I'm more than happy for them to apply a little bit of competitive pressure on Llama just as I am for the reverse to be true.

@pmarca The wealth tax proposal in the attached doc is different from what you describe in the podcast. No immediate tax if your cap gains are 80% illiquid (pre-IPO startup founders) and wouldn't apply to VC funds. Can you please explain your remarks to a fan of the pod?

References mentioned in podcast: Proposal to tax unrealized capital gains: home.treasury.gov/system/files/1… Trump 2024 platform: documentcloud.org/documents/2479… Little Tech Agenda: a16z.com/the-little-tec…

New podcast, @bhorowitz and me, 90 minutes on the Little Tech Agenda, the future of tech in America, politics, Biden and Trump. youtube.com/watch?v=n_sNcl…

@HardcoreHistory I've listened to every episode but the Alexander episode was the one I was waiting for! Thank you 🙏🏼

Happy Hardcore History Day everyone! HH #72 ""Mania for Subjugation" is now available. Hope you like it. dts.podtrac.com/redirect.mp3/t…

Such a good idea 😂😂

Can't believe @stlouisfed misleading conclusion that LLMs outperform professional inflation forecasters. How to ensure the model doesn't cheat given it's seen the answers? "Prompt it to pretend today’s date is at a certain point in the past"🤦 …deralreservebankofstlouis.cmail19.com/t/t-e-zukiitd-…