ed catmull basically solved one of the hardest organisational problems in creative work: how do you keep very high standards of taste without turning the people who hold that taste into gatekeepers that slowly strangle new ideas? his answer is elegant and I think underrated: you separate the people who set the bar from the people who make the final calls.

wow they’re really committed to this academy thing

a surprising amount of what you want is probably available, mostly because very few ppl ever decide what they want then move toward it with clear intent you’d be surprised how far that alone gets you. have a clear vision, commit to it like it’s non-negotiable, then apply a stupid amount of effort + iteration until reality starts responding two things i see people do all the time that kill this: 1. doing X so they can eventually do Y, instead of just doing Y (ex: i’ll fix my portfolio then apply, i’ll learn a bit more then start, i’ll get in shape then sign up). you mostly get ready for the thing by doing the thing 2. being way too afraid of getting it wrong. whether you succeed on any individual attempt is almost secondary. decisiveness gets you information faster, gives you more cycles to try again, and keeps momentum alive most people massively overestimate how locked in reality is. you don’t need permission, you can just start doing things. that’s what “high agency” actually is

stacked lineup

i'd pay for this

delusion is the only remaining moat

everything reminds me of him..

why spend 3 minutes enjoying a song when you could be done in 2:18

only 570 views on this somehow... another proof that view count remains a really bad proxy for whether something is worth your time

banger from the salesforce social team smart to show they understand why ppl find it funny & just let themselves be part of the joke instead of trying to drag everything back toward dreamforce messaging @Benioff liking a bunch of the memes helps too since now ppl are reposting “lol benioff liked this” and the same joke gets another round of distribution still have no idea what salesforce is doing but pretty hard to miss that dreamforce is happening rn

one thing ppl underrate is that a head of content/social is basically running an internal venture fund where the scarce capital is attention everyone in the company is pitching you for some of it (for their product launch, customer story, partnership, etc.) and you can fund all of them a little, but you end up with a very diversified portfolio of medium bets the best heads of content i've worked with make very few bets & have much more of a concentrated investor brain than an editor brain they have a thesis on the 3-4 narratives they think can become huge, keep putting attention behind them, double down when one starts working, help the teams make the underlying ideas sharper, find better formats / proof / distribution, kill stuff that distracts from them and basically keep steering the company back toward the same few bets similar to good VCs, a lot of their edge comes from being able to see things early, knowing which idea / format / shift is worth investing in before everyone has the same take on it the day-to-day makes this really easy to lose btw because there is always another post to ship and another team that wants their thing promoted, so without someone having strong opinions & making those allocation decisions the calendar just fills itself the failure mode is a year of individually defensible allocation decisions somehow adding up to a brand nobody strongly associates with anything

closed weights meet heavyweight

Last month I wrote about how we can build a positive and safe future for everyone: meta.com/thefutureisforevery… Every lab has the responsibility and incentive to move at the pace required to train its models safely, and the ability to take its own actions to ensure that happens. The reality is: - People won't want to use agents that are misaligned with them and that don't do what they ask, so labs have a strong natural incentive to make their models more aligned. There is a lot of debate about slowing progress on capabilities until alignment catches up. My view is that trust and alignment are quickly becoming the most important capabilities that will differentiate agents and models. Any lab that doesn't focus on alignment will fall behind. - Labs face significant liability if their models cause harm, so they have a strong incentive to prevent this as well. Meta delayed shipping Muse for several months to focus on safety and security. We didn't call for everyone else to do this before we would. We just did it as part of our day-to-day work because it was clearly the right thing for people and for us. I'm proud of the security foundations we've built. - Engaging independent evaluators and advisors is industry best practice. MSL already does this today in several areas because it helps produce better work. Other labs can just do this too. In general, it would be helpful for there to be a larger and more diverse ecosystem of evaluators. - Committing the significant majority of compute towards serving people rather than racing towards recursive self-improvement is one of the best ways to ensure we develop this technology safely. Meta has made this commitment and other labs can do this as well. I believe the key to building a positive future for everyone is maintaining the right balance of power. This is within our power to do.

this is a surprisingly good map of what actually makes an achievement feel good there’s an internal side: how hard something was for you & how proud you are of what you did and an external side: whether anyone notices, and more importantly whether the person noticing is someone whose judgment you actually trust getting a ton of praise for something you privately know was easy / mediocre feels good only for about 5 minutes. same, doing something incredibly hard that you’re genuinely proud of can feel great on its own, but there’s still something very different about having the exact person you wanted to notice it say “holy shit, this is good” the top right is the magic combination: you pushed yourself further than you thought you could, you respect the result, someone notices, and you deeply respect the person noticing this is also multiplicative (effort × pride × recognition × weight of the person’s opinion), so if one variable gets close to zero the whole thing collapses (which explains why some huge wins feel flat)

don’t confuse motion with progress tho you can work insanely hard for years and still end up basically where you started richard hamming has this “drunk sailor” analogy: take n random steps left and right and most of them cancel each other out, so you only end up about √n steps from where you started being busy can feel like you’re going somewhere because there’s constant evidence that something is happening (working, shipping, learning, meeting people, taking on more stuff), but a surprising amount of that effort just cancels itself out if you keep changing direction probably even truer for ambitious people because they’re rewarded for motion (the “you can just do things” mentality) while also feeling the constant pressure to keep doing better / seize the latest opportunity / not get left behind, and end up chasing a different life every few months been asking myself a lot lately whether what i’m doing rn makes my life 3 years from now easier / better / more interesting in some actual way sometimes the answer is very obviously no and i’m just busy

sf is so back (inside)

not growth related but this post had me thinking about what i’d look for in a new media person because “good content person” feels way too broad for the moment we’re in if i were hiring for new media today, the person i'd want is usually strong in these 4 areas: >editorial judgment >format instinct >distribution awareness >taste under pressure someone that knows what is worth saying in the first place, knows whether something should be a post, a clip, a thread, etc., understands that publishing is not the finish line and can move fast without defaulting to whatever language / aesthetic the internet is overproducing this month a lot of people have 1 or 2 of these but very few have all 4

why haven’t we seen more breakout work from people using ai? so far, ai hasn’t turned “mid” writers into great ones. great writers are still great (now just faster). bad writers are still bad (now just louder). but we haven’t yet seen a “mid” writer who drops something undeniably world-class because of how they used these tools, even though that should be possible given the effort that’s gone in. so something’s still somehow missing. maybe it’s to do with the tools or maybe it’s to do with the process. my feeling is today’s systems are probably already “good enough” to support exceptional work, but only in the hands of someone with obsessive taste, discipline, and vision, someone who treats ai like raw material, experiments obsessively, and brings the actual soul of the project. if that’s true, the limiting factor is our taste, processes and how far we’re willing to push. i’m mostly just curious to see what that new kind of power user looks like when they finally show up.

I think about this a lot.