casey hamlyn

This speech hits because of the hidden psychology underneath it, not the shame. 200 years ago a 15-year-old didn't need motivation to build a cabin. The weather was the motivation. Fail to build it and you freeze. The feedback loop was immediate, physical, and binary. You either did the thing or suffered direct consequences measured in days, not decades. Modern teens face the opposite problem. The consequences of not doing homework are real but abstract, delayed by months or years, and invisible at the moment of decision. Your brain literally cannot treat "this affects your GPA which affects college which affects your career trajectory in 8 years" the same way it treats "build shelter before November or die." The prefrontal cortex that processes long-term planning doesn't finish developing until age 25. You're asking a half-built brain to resist every dopamine trap ever engineered by billion-dollar companies in exchange for a reward it can barely conceptualize. The teacher is right that the task is easier. What he's accidentally describing is that motivation gets harder in proportion to how safe your environment becomes. Survival pressure is the most powerful productivity system ever created, and we removed it. The replacement is supposed to be discipline, but discipline is a skill nobody teaches. They just say "you have no excuses" and expect the kid to figure it out. The 15-year-old in 1825 didn't have more grit. He had fewer options.

A man just pled guilty to stealing $8 million from Spotify, Apple Music, and Amazon. His songs had billions of streams. Not a single human being ever pressed play. Here’s how the math worked. Smith had 1,040 bot accounts. Each one streamed roughly 636 of his AI-generated songs per day. That’s 661,440 fake streams every 24 hours, at about half a cent per stream: $3,307 per day, $99,216 per month, $1.2 million per year. For seven years, from a suburb of Charlotte, using songs made by an AI tool called Boomy that he spread across thousands of tracks so no single one triggered fraud detection. The platforms paid him the way they’d pay any artist. Their systems couldn’t tell the difference. That makes this sound like one guy’s scheme. The number you should actually pay attention to is different. Spotify paid $11 billion in royalties last year from a shared pool. Every subscriber’s $10.99 goes in, and every stream determines how the pot gets split. When Smith’s bots played 661,440 times a day, each play shrank the per-stream payout for every legitimate artist by a fraction of a cent. The 100,000th highest-earning real artist on Spotify made $7,300 for an entire year. Smith pulled $1.2 million annually from the same pool with music no human chose to hear. Now consider the flood. Deezer receives 60,000 fully AI-generated tracks every single day. 39% of all new music on the platform. In January 2025: 10,000 per day. September: 30,000. November: 50,000. January 2026: 60,000. 85% of streams on those AI tracks are fraudulent. Spotify removed 75 million “spammy songs” last year. Over 100,000 new tracks hit streaming platforms daily. The first criminal prosecution for streaming fraud took nine years to land a guilty plea. The AI tools that powered it have gotten orders of magnitude cheaper since Smith started. The DOJ proved this is a crime. They haven’t answered how many people are running the same playbook right now without tripping a single wire.

The player on the right is 6’8 and looking up in amazement at the player on the left, who’s listed at 7’9. Probably the first time he’s ever felt short.

$APP divested its gaming studios when they realized that auction liquidity provides a higher fidelity signal than 1P gameplay data. Their Axon algorithm does not need to know user behavior inside a game because it learns directly from the realtime bid density and conversion rates across the entire network. This proves that market depth creates a defensible moat that is superior to the capital heavy model of content ownership. They apply this same efficiency to their technical infrastructure by refusing to train proprietary foundation models. Management views large language models as a commodity utility so they simply integrate the best external models to solve specific bottlenecks like fraud detection and ad creative generation. This allows them to execute faster than competitors who waste capital trying to build models that quickly become obsolete. This operational discipline relies on a contrarian hiring strategy that explicitly avoids high profile AI researchers who prioritize academic publishing. They maintain a lean team who focus exclusively on shipping production code which generates profit margins per employee that far exceed their big tech competitors.

Mike Tomlin went 68-48-1 since 2019 with 5 playoff berths using: • Ben Roethlisberger in final years • Mason Rudolph • Duck Hodges • Mitch Trubisky • Kenny Pickett • Justin Fields • Russell Wilson • Aaron Rodgers in final year(s) Won’t find many HCs who can do that.

This kid is only 7 years old 🤯 (Via @HardwoodReportX 🎥)

John Cena talks about why he smiled as he tapped out in his final match: “So now I'm out there. They hate the guy I'm working with. They universally cheer me, that is f*cking rare. And for the last 5 minutes, everything I preach about story and drama and having a conversation with the audience, the ones I love are in the front row. I know my colleagues are watching on the monitor back there. We're just in a sleeper hold — So, as I essentially take my last breath, I have struggled. If you think of somebody, the natural cause of scenario or however we picture loss in our life, all of us have been through it. They struggle, they struggle, they struggle, they hang on just long enough to make sure to say goodbye to everybody that's been meaningful in their lives. And that whole day was so many unbelievably vulnerable, meaningful conversations. And then you realize, I've connected with everybody I love. Physically, I feel great. I think it's time to take that last breath. And that's that, I hate to keep going back morbidly to obituaries, but like this person died peacefully.” (via What Do You Wanna Talk About)

He is in first place right now for special teams but needs your help to keep him there. Retweets count as votes. Help him out. #ProBowlVote  Ben Skowronek #ProBowlVote  @BennySkow

Chris Boswell is not currently in the top 10 leading Pro Bowl vote-getters for kickers and that needs to change. Retweets count as votes so make sure to RT this! #ProBowlVote #ChrisBoswell #Steelers #NFL

Ben Skowronek is currently the leading vote-getter for special team player. Retweets count as votes. #ProBowlVote @BennySkow #Steelers #NFL

Retweets count as votes. Help me out Steelers Nation. @steelers #ProBowlVote Ben Skowronek #ProBowlVote @BennySkow

Incrementality Factors plotted from 72 test results. Google Brand shows the lowest incrementality % on average, Meta ACQ the highest, and Google NB the widest spread. Discovering your brand’s factor is one of the most critical data points to inform budget allocation decisions.

Duke led by 14 with 8:03 left. Duke led by 9 with 2:06 left. Duke led by 6 with 34 seconds left. Duke led by 3 with 25 seconds left. Duke led by 1 with 19 seconds left. Duke lost.

My article in Forbes: AI Shopping Agents Are Here: They Will Reshape Retail And Advertising buff.ly/411HFJe

Mike Tomlin telling Mike williams “Your a Steeler now “ Got me hyped 😭

LVMH buying indirectly into Moncler. Language strikingly similar to Arnault comments on Rupert & Cartier. Whether €15 or €75 billion cap, everyone in luxury speaks to $LVMUY first. And downturn or not, it's still in acquisitive mode. $CFRUY $HESAY $EL

My friends at Google are getting an earful from me right now. Demand Gen will be interesting and I’m sure it’ll work reasonably well but it’ll be because they are packaging great inventory with not great inventory. And then marketers will have to buy upper and mid funnel YouTube campaigns to hack the black box and find incrementality. Google’s product marketing team not even pretending to care about advertiser controls anymore.

Abercrombie & Fitch Q2 results: 💵 Total sales: +21.2% 🪙 Comp sales: +18% 👕 Abercrombie sales: +25.8% 👖 Hollister sales: +16.7% 📈 Net income: +130.5% What is there to say? They've done it again! Exceptional performance off the back of an exceptional performance last year!

Retail media update: -Walmart's ad business grew 26%. Was last reported to be $3.4 billion. -Target's ad business saw "double digit growth" and will end year in "high teens." It's "valued" at $1.5 billion. Lotta money floating around.

Jalen Brunson: "How many?" Rick Brunson: "Till I tell you to stop. Go"