Offshore
Photo
Illiquid
Limit up today. Just crazy action everywhere.
tweet
Limit up today. Just crazy action everywhere.
Furuya Metal (7826.T) is a niche precious-metals tech leader, critical „pick and shovel“ inputs for electronics & data centers.
One of the few global specialists that can refine and process iridium & ruthenium. Market shares up to 90% in niches. Targets near-3x revenue by 2030. https://t.co/x6B23zVak3 tweet
Offshore
Photo
Michael Fritzell (Asian Century Stocks)
RT @kos_data: 🇷🇺 Russian President Putin net approval rating in Europe:
Only fans:
🇷🇸 Serbia: 37
Top haters:
🇺🇦 Ukraine: -98
🇩🇰 Denmark: -94
🇸🇪 Sweden: -94
(Poll: Gallup International) https://t.co/RotlyhIDzc
tweet
RT @kos_data: 🇷🇺 Russian President Putin net approval rating in Europe:
Only fans:
🇷🇸 Serbia: 37
Top haters:
🇺🇦 Ukraine: -98
🇩🇰 Denmark: -94
🇸🇪 Sweden: -94
(Poll: Gallup International) https://t.co/RotlyhIDzc
🇺🇸 US President Trump net approval rating in Europe:
Top fans:
🇽🇰 Kosovo: 27
🇷🇴 Romania: 11
🇲🇩 Moldova: 10
🇲🇰 N.Macedonia: 2
Top haters:
🇩🇰 Denmark: -84
🇸🇪 Sweden: -80
🇳🇴 Norway: -79
(Poll: Gallup International) https://t.co/XCAXSStNBt tweet
Offshore
Photo
God of Prompt
RT @alex_prompter: Claude Opus 4.6 just became the most dangerous competitive intelligence tool on Earth.
I reverse-engineered my competitor's entire strategy in minutes.
Found their pricing, positioning, weaknesses, and future roadmap.
Here's the prompt (use responsibly):
---
"Conduct deep competitive intelligence on [COMPETITOR NAME]:
COMPANY OVERVIEW:
- Founding story and key milestones
- Leadership team (backgrounds, previous companies)
- Funding history (rounds, investors, valuations, burn rate estimates)
- Employee count and growth trajectory (check LinkedIn headcount)
- Office locations and expansion patterns
PRODUCT DEEP-DIVE:
- Complete product catalog with descriptions
- Pricing tiers (current + historical changes)
- Feature comparison vs top 3 alternatives
- Technology stack (from job postings, tech blogs, BuiltWith)
- Recent product launches (last 12 months)
- Roadmap clues (from: job postings, conference talks, patent filings, customer surveys)
MARKET POSITIONING:
- Target customer (size, industry, characteristics, job titles)
- Ideal Customer Profile (ICP) based on case studies
- Messaging and positioning (analyze website, ads, content)
- Brand voice and personality
- Key differentiators they claim
GO-TO-MARKET STRATEGY:
- Marketing channels (paid, organic, partnerships)
- Content strategy (blog topics, frequency, engagement)
- Sales approach (inbound vs outbound, PLG vs sales-led)
- Partnership ecosystem (integrations, resellers, tech partners)
- Event presence (conferences, webinars, sponsorships)
CUSTOMER INTELLIGENCE:
- Review analysis (G2, Capterra, TrustPilot - what do users love/hate?)
- Common complaints (from Reddit, Twitter, support forums)
- Feature requests and gaps (from public roadmap, user forums)
- Churn signals (Glassdoor reviews, customer testimonials that stopped)
STRATEGIC VULNERABILITIES:
- What are they bad at? (based on reviews, hiring patterns)
- What markets are they ignoring?
- Where are they overextended?
- Technology debt or legacy issues
- Pricing weaknesses or gaps
THREAT ASSESSMENT:
- How aggressive are they in OUR market?
- What would it take to compete effectively?
- What could they do that would hurt us most?
- Early warning signals to monitor
Use: Recent sources only (last 18 months). Prioritize primary sources (their blog, official announcements, verified reviews). Flag speculation vs confirmed facts. Include URLs for verification."
---
tweet
RT @alex_prompter: Claude Opus 4.6 just became the most dangerous competitive intelligence tool on Earth.
I reverse-engineered my competitor's entire strategy in minutes.
Found their pricing, positioning, weaknesses, and future roadmap.
Here's the prompt (use responsibly):
---
"Conduct deep competitive intelligence on [COMPETITOR NAME]:
COMPANY OVERVIEW:
- Founding story and key milestones
- Leadership team (backgrounds, previous companies)
- Funding history (rounds, investors, valuations, burn rate estimates)
- Employee count and growth trajectory (check LinkedIn headcount)
- Office locations and expansion patterns
PRODUCT DEEP-DIVE:
- Complete product catalog with descriptions
- Pricing tiers (current + historical changes)
- Feature comparison vs top 3 alternatives
- Technology stack (from job postings, tech blogs, BuiltWith)
- Recent product launches (last 12 months)
- Roadmap clues (from: job postings, conference talks, patent filings, customer surveys)
MARKET POSITIONING:
- Target customer (size, industry, characteristics, job titles)
- Ideal Customer Profile (ICP) based on case studies
- Messaging and positioning (analyze website, ads, content)
- Brand voice and personality
- Key differentiators they claim
GO-TO-MARKET STRATEGY:
- Marketing channels (paid, organic, partnerships)
- Content strategy (blog topics, frequency, engagement)
- Sales approach (inbound vs outbound, PLG vs sales-led)
- Partnership ecosystem (integrations, resellers, tech partners)
- Event presence (conferences, webinars, sponsorships)
CUSTOMER INTELLIGENCE:
- Review analysis (G2, Capterra, TrustPilot - what do users love/hate?)
- Common complaints (from Reddit, Twitter, support forums)
- Feature requests and gaps (from public roadmap, user forums)
- Churn signals (Glassdoor reviews, customer testimonials that stopped)
STRATEGIC VULNERABILITIES:
- What are they bad at? (based on reviews, hiring patterns)
- What markets are they ignoring?
- Where are they overextended?
- Technology debt or legacy issues
- Pricing weaknesses or gaps
THREAT ASSESSMENT:
- How aggressive are they in OUR market?
- What would it take to compete effectively?
- What could they do that would hurt us most?
- Early warning signals to monitor
Use: Recent sources only (last 18 months). Prioritize primary sources (their blog, official announcements, verified reviews). Flag speculation vs confirmed facts. Include URLs for verification."
---
tweet
Offshore
Photo
Michael Fritzell (Asian Century Stocks)
RT @ai: One Japanese company makes virtually all the ultrathin glass used in advanced AI chips. Nittobo controls the T-glass market and isn't adding capacity for months. We talk a lot about GPU supply constraints but the bottlenecks keep showing up in places nobody's watching. Chip packaging depends on this stuff. https://t.co/LrPF0HpyvX
tweet
RT @ai: One Japanese company makes virtually all the ultrathin glass used in advanced AI chips. Nittobo controls the T-glass market and isn't adding capacity for months. We talk a lot about GPU supply constraints but the bottlenecks keep showing up in places nobody's watching. Chip packaging depends on this stuff. https://t.co/LrPF0HpyvX
tweet
Offshore
Photo
The Transcript
RT @TheTranscript_: $LVMH CEO: Family ownership enables long-term strategic focus over quarterly pressures.
“A family group isn’t riveted to the quarterly results... we take the long-term view... we’ll cross the 50% threshold.” https://t.co/obOdqUB4dn
tweet
RT @TheTranscript_: $LVMH CEO: Family ownership enables long-term strategic focus over quarterly pressures.
“A family group isn’t riveted to the quarterly results... we take the long-term view... we’ll cross the 50% threshold.” https://t.co/obOdqUB4dn
tweet
Offshore
Photo
God of Prompt
🚨 Holy shit… Stanford just published the most uncomfortable paper on LLM reasoning I’ve read in a long time.
This isn’t a flashy new model or a leaderboard win. It’s a systematic teardown of how and why large language models keep failing at reasoning even when benchmarks say they’re doing great.
The paper does one very smart thing upfront: it introduces a clean taxonomy instead of more anecdotes. The authors split reasoning into non-embodied and embodied.
Non-embodied reasoning is what most benchmarks test and it’s further divided into informal reasoning (intuition, social judgment, commonsense heuristics) and formal reasoning (logic, math, code, symbolic manipulation).
Embodied reasoning is where models must reason about the physical world, space, causality, and action under real constraints.
Across all three, the same failure patterns keep showing up.
> First are fundamental failures baked into current architectures. Models generate answers that look coherent but collapse under light logical pressure. They shortcut, pattern-match, or hallucinate steps instead of executing a consistent reasoning process.
> Second are application-specific failures. A model that looks strong on math benchmarks can quietly fall apart in scientific reasoning, planning, or multi-step decision making. Performance does not transfer nearly as well as leaderboards imply.
> Third are robustness failures. Tiny changes in wording, ordering, or context can flip an answer entirely. The reasoning wasn’t stable to begin with; it just happened to work for that phrasing.
One of the most disturbing findings is how often models produce unfaithful reasoning. They give the correct final answer while providing explanations that are logically wrong, incomplete, or fabricated.
This is worse than being wrong, because it trains users to trust explanations that don’t correspond to the actual decision process.
Embodied reasoning is where things really fall apart. LLMs systematically fail at physical commonsense, spatial reasoning, and basic physics because they have no grounded experience.
Even in text-only settings, as soon as a task implicitly depends on real-world dynamics, failures become predictable and repeatable.
The authors don’t just criticize. They outline mitigation paths: inference-time scaling, analogical memory, external verification, and evaluations that deliberately inject known failure cases instead of optimizing for leaderboard performance.
But they’re very clear that none of these are silver bullets yet.
The takeaway isn’t that LLMs can’t reason.
It’s more uncomfortable than that.
LLMs reason just enough to sound convincing, but not enough to be reliable.
And unless we start measuring how models fail not just how often they succeed we’ll keep deploying systems that pass benchmarks, fail silently in production, and explain themselves with total confidence while doing the wrong thing.
That’s the real warning shot in this paper.
Paper: Large Language Model Reasoning Failures
tweet
🚨 Holy shit… Stanford just published the most uncomfortable paper on LLM reasoning I’ve read in a long time.
This isn’t a flashy new model or a leaderboard win. It’s a systematic teardown of how and why large language models keep failing at reasoning even when benchmarks say they’re doing great.
The paper does one very smart thing upfront: it introduces a clean taxonomy instead of more anecdotes. The authors split reasoning into non-embodied and embodied.
Non-embodied reasoning is what most benchmarks test and it’s further divided into informal reasoning (intuition, social judgment, commonsense heuristics) and formal reasoning (logic, math, code, symbolic manipulation).
Embodied reasoning is where models must reason about the physical world, space, causality, and action under real constraints.
Across all three, the same failure patterns keep showing up.
> First are fundamental failures baked into current architectures. Models generate answers that look coherent but collapse under light logical pressure. They shortcut, pattern-match, or hallucinate steps instead of executing a consistent reasoning process.
> Second are application-specific failures. A model that looks strong on math benchmarks can quietly fall apart in scientific reasoning, planning, or multi-step decision making. Performance does not transfer nearly as well as leaderboards imply.
> Third are robustness failures. Tiny changes in wording, ordering, or context can flip an answer entirely. The reasoning wasn’t stable to begin with; it just happened to work for that phrasing.
One of the most disturbing findings is how often models produce unfaithful reasoning. They give the correct final answer while providing explanations that are logically wrong, incomplete, or fabricated.
This is worse than being wrong, because it trains users to trust explanations that don’t correspond to the actual decision process.
Embodied reasoning is where things really fall apart. LLMs systematically fail at physical commonsense, spatial reasoning, and basic physics because they have no grounded experience.
Even in text-only settings, as soon as a task implicitly depends on real-world dynamics, failures become predictable and repeatable.
The authors don’t just criticize. They outline mitigation paths: inference-time scaling, analogical memory, external verification, and evaluations that deliberately inject known failure cases instead of optimizing for leaderboard performance.
But they’re very clear that none of these are silver bullets yet.
The takeaway isn’t that LLMs can’t reason.
It’s more uncomfortable than that.
LLMs reason just enough to sound convincing, but not enough to be reliable.
And unless we start measuring how models fail not just how often they succeed we’ll keep deploying systems that pass benchmarks, fail silently in production, and explain themselves with total confidence while doing the wrong thing.
That’s the real warning shot in this paper.
Paper: Large Language Model Reasoning Failures
tweet
Michael Fritzell (Asian Century Stocks)
RT @anonymous3nibrv: Stronger yen could be the catalyst.
It's be hard not to hold 'at least some' growth
as people de-risk from weak yen beneficiaries.
I'd be comfortable half-kellying 6562 Genie.
Will study a bit and do a writeup tomorrow morning.
tweet
RT @anonymous3nibrv: Stronger yen could be the catalyst.
It's be hard not to hold 'at least some' growth
as people de-risk from weak yen beneficiaries.
I'd be comfortable half-kellying 6562 Genie.
Will study a bit and do a writeup tomorrow morning.
Japan really is a two-tier market… I feel more comfortable with the growth index at this point tweet
X (formerly Twitter)
Michael Fritzell (Asian Century Stocks) (@MikeFritzell) on X
Japan really is a two-tier market… I feel more comfortable with the growth index at this point
Offshore
Photo
Javier Blas
CHART OF THE DAY: Vitol, the world's largest independent oil trader, shifts its view on global peak oil demand: higher and later.
Now, Vitol sees a peak by the "mid" 2030s (previously it anticipated "early" 2030s). Peak demand seen at ~112m b/d (previously "almost" ~110m b/d). https://t.co/67qVesRu2r
tweet
CHART OF THE DAY: Vitol, the world's largest independent oil trader, shifts its view on global peak oil demand: higher and later.
Now, Vitol sees a peak by the "mid" 2030s (previously it anticipated "early" 2030s). Peak demand seen at ~112m b/d (previously "almost" ~110m b/d). https://t.co/67qVesRu2r
tweet