Jonathan Wu @jaypwu

Uber says it has limited all employees to $1,500 in monthly token spending per AI coding tool "to responsibly encourage agentic AI adoption" (@natlungfy / Bloomberg) (Visit Techmeme dot com for the link and full context!)

Been a management consultant for 20 years. Made Partner in my 30s. Led teams of 100+ people. Run 9-figure client portfolios. Lived and worked in 4 continents. Typically, corporate IT investment would follow a common script. Capital spent on software means a shrinking payroll. As boards map out their strategies for the coming quarters, they are operating under the comfortable assumption that this way of thinking still holds true for AI. But I think a fiscal reckoning is brewing there, because within the next few quarters, the current prevailing narrative of AI as a headcount killer (which we all know is vastly exaggerated) will give way to a far more punishing reality. Instead of a clean capital-for-labor swap, executives are about to watch their IT infrastructure costs and their personnel expenses balloon simultaneously 🚀🚀🚀 It may not be fun. First, this whole idea that generative AI can operate autonomously will shatter as early deployments attempt to scale. Because LLMs remain inherently prone to hallucination and error, companies cannot simply fire the analysts; they will be forced to retain them (or hire new talent) to serve as high-vigilance editors. Furthermore, because AI makes it effortless to generate code, reports, marketing collateral, etc etc organizations will soon find themselves drowning in internal output. Managing, auditing, and securing this massive influx of AI-generated material will require an unprecedented wave of human oversight.... This will ultimately EXPAND corporate bureaucracy rather than trimming it (remember the 'Scaled Agile' saga??). Even in scenarios where entry-level automation does succeed, the math of headcount reduction will fail to balance out on the ledger. In the coming quarters, the wage differential of the AI era will trigger *severe* skill inflation. Replacing 5 mid/entry-level programmers does not result in a net savings of 5 salaries. Instead, it requires hiring a premium-tier AI architect whose single salary frequently eclipses the combined wages of the workers they replaced (plus tokens cost). Companies will trade high-volume/low-cost labor for scarce/ultra-premium talent, driving TCO UPWARD despite a leaner organizational chart on paper. Jevons' Paradox again... AI slashes the time and cost required to draft a legal brief, design a graphic, build a software feature, and therefore executive appetite for those outputs will skyrocket. Management will demand 10x the volume of data analysis or continuous product iterations. Because the corporate demand for output will scale far faster than the technology's efficiency gains, departments will find themselves forced to expand their human teams just to handle the sheer velocity of these new AI-driven initiatives. Until AI achieves absolute, unmonitored autonomy (if ever), it will function not as a replacement for human labor, but as a hyper-amplifier of it. If ungoverned, the corporate balance sheets will show that the AI boom made running the business vastly more expensive.

@odysseus0z What’s the cost must be millions but speed means it’s a winner

$200k is not a good deal to join most SF startups now Employee equity is riskier today Huge dilution from bigger rounds, more and more tricky acquisitions that exclude employees, startups not going public, secondaries getting blocked… And 1 bedrooms are $4k/mo+

So we got a multibillion dollar captive out there. Kind of. More or less. @CORGI is suddenly a massive player in the startup business insurance space. Here is a breakdown on what the company is & how its doubled its billion dollar valuation in 3 weeks. Corgi is the program administrator for the Technology Risk Retention Group, Inc. An RRG is a member-owned liability insurance company permitted to write across state lines. It's similar to an E&S carrier but only has filings with its domicile state & has materially lower capitalization requirements. TRRG is domiciled in Arizona which has a minimum paid in capital requirement of $500k. TRRG is, in turn, reinsured by Corgi Reinsurance, Inc. Corgi Re is also an Arizona domiciled entity and appears to be a captive. Both TRRG & Corgi Re are managed by Dane Management, LLC. So what is Corgi? And why is it so important?

@Sabrescokegate Bah construction is nice with storage more of a half cart bag half carry bag. Sun Mountain or Titleist Carbon or Vessel Carbon probably best for carry only weight and versatility

@Mike_kim714 Bringing the baby!

Southampton NY isn’t the easiest place to get to so I might as well get this started early! Ticket giveaway for the US open! Bonus points if you take a kid (12 and under are free)

everyone making the monthly migration back to claude from codex

Andrej Karpathy @karpathy

2 weeks ago

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

8K 11K 150K 27.4M 14K

@atrupar Just do both. No income tax on low income individuals, tax symbolically the rich

Enterprises are about to waste billions sending every query to a frontier model when most should be routed across a mix of cheaper, specialized ones. Writing first checks into founders building the intelligent routing layer.

@luagaroz EBITDA

Market top signal is near give it 6 months to blow off the top

David Sacks @DavidSacks

3 weeks ago

Was thinking about getting a $40k Royal Oak. Will get this for $400 instead. 99% savings — thanks AP!

540 497 10K 6.9M 4K

Eli Lilly makes tirzepatide with a DEA List I controlled chemical, a Group 2A carcinogen, and a PFAS forever chemical. Novo Nordisk uses yeast. Novo's Ozempic backbone is fermented at their patent-protected plant in Kalundborg, Denmark. Tirzepatide and the synthetic GLP-1s are made via solid-phase peptide synthesis of fragments, then coupled in solution. Each cycle of deprotection uses piperidine, controlled by the DEA under Code 2704. It is the core of fentanyl. Each tirzepatide takes roughly 38 piperidine cycles across its fragments. Every SPPS step uses dimethylformamide as the solvent. DMF's residual limit was set at 880 ppm in 1997. The IARC reclassified it as a probable carcinogen in 2018, but the limit hasn't changed. Cleavage uses trifluoroacetic acid, a forever chemical. There is no harmonized regulatory limit for it. Toxicology data is too limited to set a safe daily dose. Approved drugs pass the specs. Every synthetic GLP-1 is made with these chemicals. One kilogram of API generates roughly twelve thousand liters of solvent waste. Novo built the modern route. Everyone else is stuck in 1997. Tens of millions take GLP-1 every week. We engineer enzymes to manufacture chemicals. Developing a yeast in the lab is not the same as running it in production. Novo has fermented GLP-1 peptides at industrial scale since 2010.

Every app on earth should have a natural language bar that can do anything across the app Doordash Airbnb Uber

ben hylak @benhylak

4 weeks ago

so instead of saying “funky cabins within 2 hour drive” i will have to keep filling out your patient intake form

76 10 1K 448K 256

We are two weeks from oil shortages and demand rationing according to Bloomberg. Not months…weeks. Anyone that thinks this oil shortage isn’t a serious immediate crisis is delusional:

Every time I really get along with someone

PsyPost.org @PsyPost

4 weeks ago

A new study shows that individuals with similar levels of autistic traits are naturally drawn to one another. Brain scans reveal they use alternative, highly effective neural strategies to connect, challenging traditional deficit-based models of autism. dlvr.it/TSSN7t

702 2K 36K 23.0M 16K

I have a hypothesis that recruiting people from twitter with 500+ follower counts for technical roles tends to have a high hit rate on finding people who get shit done because it requires some amount of agency to do it, and that's a great proxy for other kinds of competency.