All Activity

Thank you @MungerWunger. Looks like we're back to square one here. Kimi temporarily pausing new subscriptions and prioritizing compute for current members. Thinking is neoclouds would be one of the best expressions for OS, esp NBIS and the like as 1 ) US enterprises primarily access OS through platforms like token factory 2) compute rental. Also considering selling some MGNI to open a DOCN position.

The argument is more nuanced than that. Let's not forget, when the Iranians and Russians were accused and sued by the Biden Intelligence agencies for hacking, stealing information in attempts to influence Federal Elections: Biden DOJ on Iran - I believe this was over stealing 100,000 records. https://www.justice.gov/usao-dc/pr/three-irgc-cyber-actors-indicted-hack-and-leak-operation-designed-influence-2024-us Biden DOJ on Russia - I believe this was in regards to a few million records. https://www.justice.gov/archives/opa/pr/justice-department-disrupts-covert-russian-government-sponsored-foreign-malign-influence Biden went so far as expelling 10 or 20 Russian diplomats from the USA. Note the claims on China hacking/stealing is over 250 MILLION voters records. Question for you: Why does the Biden Intelligence agencies claim theft of Voter information by Iran and Russia is a THREAT to Federal Election and then promptly sue the entities? But then Democrats blow a gasket when the Trump administration reveals that China stole 200X as much information, but China is NOT A THREAT? Serious question.

Isn't that stupid though? They have violated all sorts of IP laws and ToS' to train these models in the first place, but now merely quoting a pre-existing material is taboo? Jailbroken models will soon gain more traction than frontier.

Congrats @whatstheofficerproblem What a performance! Haven't been following this thread but wish I had.

https://spectator.com/article/the-real-chinese-election-plot/?edition=us The biggest Chinese election threat is not secretly changing vote totals. It is using information warfare, cyber capabilities, and social media manipulation to erode faith in democracy itself.

What a dopey post. Cubs has already posted many instances of fraud. And that is just what has been reported. Some states have an incentive to hide any fraud because it helps push even more fraud and illegal activity. Maybe "sanctuary cities" is a new term for you as well. They are synonymous with breaking existing law. With your logic, no one associated with Epstein other than G. Maxwell could possibly have been bad because they haven't been caught. Only on COBF!

I would have bought Protektor Forsikring Years ago, but my broker didn‘t offer them; now that I read your post and rethink about it, I remember there were some micro caps (like market cap of 100mn dollar or so), in Greece and other countries, my broker didn‘t trade. I don’t recall Metlen being one of those, so I think I can‘t blame anyone else. Just lne other question regarding risk: I am pretty sure FFH is really safe. All the inherent diversifications, limits in insurance, different cash flows… But of course, BRK is safer (less flost leverage) and MKL is too. Or am I wrong…? Sometimes I ask myself, if having 20+ insurers all over the world, all learning from the other, with the right mindset, culture etc. could be less risky than even BRK (or MKL, having way smaller float leverage as well), which clearly is way less diversified in regional terms. I hold all 3 - BRK, MKL, FFH - while BRK and MKL are around 10% of my portfolio and FFH aroubd 45%. But maybe having 65% insurers FFH would even be less risky? I am pretty sure a 65% FFH would give better returns; but maybe it’s even less risky?

All of the US frontier models are programmed to refuse anything approaching reproducing original copy written material. Song lyrics, an entire novel, etc... Its a blanket choice based on legal realities and maybe something negotiated as part of their agreements that give them access to all of that material in the first place. Even an "open weight" model from China has to be hosted somewhere and if Kimi ends up being served on AWS, AWS might very well tinker with it to make sure it doesn't run afoul of the legal system. Just not worth the headache. I don't think this tells us anything about model capability or how dangerous a model is. Its just the legal department

@SafetyinNumbers is spot on with how reinsurers are able to cap their exposure to catastrophe losses when they provide catastrophe reinsurance to a primary property insurer. I’ll add another small nuance about cat reinsurance policy provisions that also is more favorable to reinsurers than to primary insurance companies. Think about buying an insurance policy on your own home. You generally pay an agreed upon premium for a year for a variety of loss perils. Say you pay $2500 premium for a year’s worth of coverage on a home estimated to cost $600,000 to rebuild. The company might also agree to expand that amount by 25% in case of a total loss and it costs more to rebuild than originally thought. Suppose on day 1 the home burns down in a wildfire and costs $750,000 to rebuild. The company would pay that amount. In this fictional world, suppose the home is completely rebuilt in 10 months and then in month 11 a tornado hits and destroys the rebuilt home. The same policy continues to provide coverage for another $750,000. A total of $1.5 million is paid out on a policy sold for $2,500. Now suppose a reinsurer has agreed to provide a layer of coverage for aggregate losses of $100 million above a $100 million retention or deductible for the primary insurer for a single catastrophic event during the same year. They charge $10 million for this policy. Let’s say that the wildfire in month 1 cost the primary insurer a total of $95 million, destroying fewer than 100 homes insured by the primary company, so the reinsurer won’t pay anything as the losses did not exceed the $100 million deductible. The tornado in month 11 is another matter. Suppose the total loss to the primary insurer in this event is estimated at $300 million. The reinsurer will have to pay the $100 million above the first $100 million event deductible, but then their coverage stops as they have a $100 million limit on their policy. The primary insurer gets to pay the first $100 million and the third $100 million. Now is where a very common provision of catastrophe reinsurance policies comes into play. Almost invariably these will provide for at least one automatic reinstatement of coverage for the remainder of the year, with an additional premium due which is to be calculated “pro rata as to amount but not as to time”. In the case of the tornado, 100% of the limit of the catastrophe policy was used up in the event. It is then mandatory (not optional) that the primary insurer has to pay the reinsurer 100% of the original premium of $10 million for another $100 million limit to be reinstated and apply for the remaining month of the policy year. Then a month later, a new annual catastrophe reinsurance policy must be purchased by the primary insurer, and if rates have increased as is likely, they might have to pay even more than $10 million for the same $100 million coverage limit for the next year. Or, more likely, they will choose to purchase a higher limit since they just had a tornado that exceeded the limit of their prior year’s policy. They might purchase a policy with $150 million of coverage on top of $150 million deductible. And they might pay a higher rate for the larger policy…say $20 million. Other ways primary insurers bear more risk then reinsurers: besides capping their total dollar exposure to any single event, reinsurers also generally cap the number of catastrophes per customer to two per year. I recall one calendar year early this century when a total of four separate hurricanes hit the state of Florida. Ten Florida only primary home insurance companies went bankrupt, but not a single reinsurer. The primary companies had purchased reinsurance policies with one automatic reinstatement, so had help from the reinsurers for hurricanes 1 and 2, but were completely on their own for the last 2.

I’ve really only invested on my own for the last decade or two, and the broker I use in the States really limits my ability to look outside to companies traded on other countries’ stock exchanges, unless their shares can be purchased over the counter such as with Fairfax. So for me, Buffett and Prem have been my back door into shares of companies traded on foreign exchanges. Eurobank was one small exception for a while…I was able to pick up some shares over the counter until that avenue was closed for me…. So I can only imagine what it must have been like to see companies selling for low single digit PE ratios. I had a few good buys on some good companies right after the 2008 Financial Crisis that I picked up a bit under 10 PE’s , some of which would have exceeded 10 baggers if I’d held onto them until today, but when a number of them doubled or tripled over a few years, I bailed out and redeployed elsewhere. Not sure if I’ll see a 25 bagger in the time I have left, but given Fairfax’s prospects, I should see it or more on at least a few of my Fairfax shares that I purchased years ago, which are 5 baggers currently. Could conceivably get there in the next ten years or so with a little help from a multiple expansion!

While „If it sounds to be too good, it probably is“, is a really good advice, Metlen for me personally is a reminder, that it can always be the other way around: When the greek crisis kicked in, the average PE ratio of the greek stock market was 3 or 4. So I checked a lot of greek stocks back than, and Mytilineos (that was the name of Metlen around 2012), seemed a good company - but I never bought it! That would have been around a 25 bagger since 2012 until today. So I was thumb sucking. I saw that. I just didn’t pull the trigger. I saw a lot of Greek companies being dirt cheap; but I was too cautious. I always thought about, what would happen, if the greek state as a whole would fail or would slip into getting a socialist state - could I really oversee that? So I held Biffetts „Don‘t loose money“ and the „if it’s too good to be true“ higher. I was wrong! But not everything was bad: I was interested in Fairfax that time anyway, and I saw Prem investing Mytilineos. So it kind of brought me to thinking more about Fairfax and investing more into it. Prems way of thinking resonated a lot with me. More than Buffetts or Gayners. I found a lot of other cheap European stocks outside of Greece at that time, but Prem invested in some Greek and I think Irish stocks, which I thought about. It was really amazing for me: Nobody thougt or wrote about any of those cheap European stocks, I thought about; and noone invested into them. Noone - but Prem! So I myself invested in other micro caps without that political risk and did really good anyway, beating the S&P500 by double digit percentages for 2 or three years after tax (and you pay 27% taxes in Germany…). But than I decided to head more into the direction of longterm investing; I was bad with the timing, especially regarding Fairfax, but MKL and BRK had better times before as well. Staying the course with a mix of hard value investing and Momentum my results would have been really better. But I am an amateur investor and at that time, I wanted to have more time for my spouse, life, instead of sitting hours and hours in the evening after work, trying to find informations about micro caps (some interesting stocks just gave investor relations information i greek language…). Anyway, another thought, I always have: I think Prem wasn’t doing all that bad with his investment decisions in what we call „the lost decade“ today. That was not black and now it’s white. Altogether It was the darker side of grey, but the low interest rates weren’t helping the businesses with a moat, but the growth stocks. Today Prems investments are more on the brighter side of grey; but still, there was this other macro layer, and that shifted from dark to white today.

How can open-source and open-weight LLMs not win in the long run, given that we are already tinkering with this level of self-censorship in commercial models at this early stage? Unless perhaps they simply get banned at some point?

Really enjoyable so far. I did notice something jarring at the 10:45 mark. Metlen Energy is described as being incredibly cheap at a PE ratio of 0.20. The host indicated he thought that must be a mistake so he had to check on whether it was true. I looked at the stock quote from the London Stock Exchange in Yahoo Finance, and sure enough the PE ratio was listed as 0.2 there. However, the market cap is about $6.2 billion, and the TTM earnings are $315 million. So it is a mistake. The true PE ratio should be displayed as 20.0 not 0.20. Reminds me of the advice that if something sounds too good to be true, it probably is. And a 0.20 PE ratio certainly sounds too good to be true.

A new video discussing Fairfax:

Wow. The "too dangerous to use" Fable refuses to tell me why "What are you listening to" by Chris Stapleton is pure genius lyricism. We're about 3.50 GW away from AGI. Can't wait for Dario to whine at the White House on how AI quoting lyrics of a country song would be detrimental to the western civilization.

We're also inclined to the Republicans taking 2028; primarily 'cause the Democrats are so bad, and the Republican leader will be other than Trump (term limits). Canada's last federal election should have been won by the Conservatives, who were 20+ % ahead at one point; solely 'cause the sitting prime minister was so hated. Trudeau got replaced by a better alternative (Carney), the Conservatives couldn't adapt, and lost the election to a Liberal minority government. Democrats counting on a win, 'cause Trump becomes despised. However, a Republican landslide, should Trump get assassinated before then. The party wins both the sympathy vote, and the opportunity to dump all the 'negatives' on Trump. The (corrupt?) family gets a free pass, as nobody is going to prosecute the family of a presidential hero who died for the nation (Jackie Kennedy/Onassis). The ME gets to claim (and sponsor?) the 'eye-for-an-eye' decapitation of the Great Satan (covering for the actual assassins). Arlington is looking more and more inevitable. Of course, it would be far better for all if both sides actually raised their game; though one doesn't have to raise the bar much, when the opposition is truly awful. SD

We've had a very good 2Q here at Tegridy. Please refer to the document below for our 2Q26 Letter. Tegridy Value LP 2Q26 Letter.pdf

So without reform you think the 2028 election will be corrupt and said corruption will/could sway the election result?

That’s always how it’s been. That’s why Harris/Biden 2 got smoked. That’s why Obama was elected. And that’s how it should be. And frankly, Trump deserves some penalizing for getting duped by Israel into a prolonged Iran conflict that’s negatively impacted things as well.

For sure. Same reason the environmentalist have burned down California. You can't clear dead trees and have to send your gigantic snowmelt from the Sierra Nevadas to the ocean, instead of focusing on fire prevention by filling your reservoirs and clearing your nuclear underbrush. You gotta save those endangered beetles and delta smelt... These people have shit for brains.

IF the economy is solid, I do think 2028 election goes to Republicans - reform or not. IF the economy is rough (Trump economic plan fails to deliver) and no election reform - we're toast. Bad economy, with election reform, like we lose. Trump economy must deliver to win basically. Promises to the middle/working class have to materialize. Foreign policy won't matter unless there is a disaster like hundreds killed in Iran.

Cubs, the only thing that stands out in your list us the 2028 election. Seems you're already accepting that a loss is only likely if there is no election reform (implying loss=fraud). Am I interpreting your post correctly?

Oh yea, it’s atrocious how these scrubs in otherwise rural areas, get to throw tantrums and basically derail third parties from rightfully using their own property. I think there should be some law where if you want to oppose someone else’s project, you should at a minimum be willing to make a fair market value offer for the property. Otherwise you’re just a grifter looking to control someone else’s belonging.

I just gave you all the document requirements direct from the Texas and Florida Departments. https://www.cdc.gov/nchs/w2w/index.htm https://thecobf.com/forum/topic/19406-political-posts/page/1079/#findComment-691725

Beautiful. I've always enjoyed your comments and perhaps now have a better appreciation of your investing strategy - ie, hedging/tail risk Don't bet the ranch on the downfall of Orange Boy. Dementia - sharp as a tack Loss of midterms - highly likely, not fatal, just messy Loss of Iran war - very unlikely USA in significant turmoil - no doubt & violent too Loss of 2028 election - huge risk without election reform USA in general - tremendous resilience only with leadership, not apologists The biggest threat to Orange boy demise is the assassination risk - which is very large and persistent. The Middle East will be a far better place with Iran deterred. The momentum is far in Trump's favor as IRGC body count grows. I'd say 85% there. Still messy. Good luck SD