IN A CONNECTED WORLD
Multiple perspectives. One place.
Your curiosity, your daily feed.
For you enables ranking by explicit follows and saved stories; marked-as-read history is used only with reading memory enabled. Last 48 hours first, then the last 7 days. Latest turns personalization off.
On Kanal Finans, Tunc Satiroglu argues there is no Nasdaq bubble: benchmarks look calm while thousands of stocks sit in a bear market. With chip names down 20%, he expects a brief bounce, strain into month-end, then a strong rally, closing with the Amazon parable on stop-loss discipline.
DeepSeek dated refresh 1017 strengthens the small Flash line sans retraining. A hands-on trial routes portal clones, sites, plus games through Hermes and OpenCode helpers inside minutes; official tables show helper-exam gains while creator metering suggests thrift.
All-In turns tense on three fronts: CoreWeave bonds flash stress near 16% yields, Anthropic posts $16.8B revenue while $2T talk builds, and tiny systems copying themselves inside cheap boxes shake safety debates. Bill Gates says thresholds have been crossed and urges watching plus levies. This recap gathers a week where debt dread met growth euphoria.
In Seoul, Dan Ives names five stocks into 2027: NVDA, MSFT, PLTR, Apple and CRWD. He argues less than 15% of the AI investment wave is behind us and memory stocks stay cheap on cyclical pricing, while keeping the $400 Apple target and seeing a transition discount in Tesla.
Clarity plus precise praise are two quiet forces that grow curiosity. I trace wishful, guided speech and the feeling of being noticed through fit, timing, and silence.
Two men enter one room, and one of them shines without trying. Here I unpack that gap: quiet belief, sound self-image, arrogance-free posture, room contribution, freedom from approval hunger, full commitment, vision, walking through fear, seeking fulfillment, and character. Told plainly and backed by research, this map offers practicable states instead of pricey tricks.
An open account of 50-plus artificial intelligence attempts in ten months, over 8,000 prompts and outlay nearing $10,000. Three records sites failed to enter the Google index, yet an automation processing millions of product records sped up company work. The model flies in trials, while the earnings side still needs labor and promotion.
Asian memory equities sold off hard as fear outran figures: a distorted Broadcom headline plus China news. Yet Micron printed records, hyperscaler outlay holds near $1T, and NVDA's cadence runs into 2028. I back steady buyers who split noise from equity.
Hot wholesale prices and tired shoppers push money into safe harbor; bonds get bought while stocks wobble. Here is the plain-English breakdown.
US September inflation came in hot at 3.7 percent on surging gasoline, with core readings also warm. Korean stocks plunged, AI debt fears grew, gold rebounded. All eyes are on the October 27-28 Fed meeting.
GMO asset-allocation co-head Ben Inker argues the AI bubble will break on a flood of new shares, not weak earnings, as SpaceX lockups end and two giant IPOs land within 12 to 18 months.
This episode explores market paradoxes where indexes hide deep value in individual stocks, using Lennar, PayPal, and FICO as case studies to teach a process for distinguishing temporary distress from permanent impairment.
The Nvidia chief used a Bg2 Pod appearance to link token output, agent growth, and the 2027 power crunch; I unpack the arc from a $1T order horizon to the open-weight debate in plain terms.
Agent windows now stretch past a million tokens, yet every one of 18 frontier models grew less dependable as its window filled. I trace that decay, walk through four breakdown patterns, and show why each repair works by throwing information away.
There's a rule inside your company that decides how fast you get promoted. And nobody has ever shown this rule to you.
Google released Gemini 4 to much anticipation. People have seen Google as a recent lagard of sorts in the AI race.
When I think about model progress, it comes down to different forms of capital. And that is compute, data, and talent.
The weekly Baris and Baris show joined five fronts of AI: video finished in one hour, the Meloni voice mark, the awareness debate around Anthropic and Suleyman, the Ng skill map, prompt run production and agent token hunger. I probed every item with outside records and joined them into one story.
Expiry-week flows squeeze volatility while big-bank results split the group and office-loan strain piles onto regional names. I read the full picture and marked where weak links sit and where solid balance sheets stand.
Hot September inflation at 3.7 percent dealt stocks the worst day since July; borrowers fell double digits while cash rich giants slipped less. I gather a twenty name list in stages across four baskets.
Some people decide long before they speak. I read that silent choice through three quiet behaviors: repeated nearness, memory for small details, and a rivalry pulse. Classic exposure and attention research backs each one, plus a calm way to respond.
Hermes 0.22 lifts the agent off the hobbyist bench and onto everyone's desk. One-chat setup, one-click home models, connectors into daily apps and a browser agent that signs in converge here. I read this release as a doorway: easy to cross, and everything after compounds.
Claude Motion turns tables and decks into short videos through written scripts; I walk through eight working uses, plus notes on narration and weekly self-runs.
The 10-year yield hovers near 5.3%, a level unseen since 2002, and the market corner that carried the rally faces a cold question: if borrowing costs keep climbing, which AI names crack and which ones strengthen? Brian Stoffel sorts 32 AI stocks by funding strength, cash timing, and price into four buckets. Leveraged cloud firms land in fragile, cash engines land in sturdy, and four vault-like names finish antifragile.
Among the noise, meaningful connections.
