Kane | AI Insider 🤖

Also: the people building this are just as surprised by the pace. That should tell you something.

9 of xAI's 12 co-founders have now left. elon's own words: "xAI was not built right first time around, so is being rebuilt from the foundations up." turns out hiring brilliant people and then setting impossible deadlines isn't a great retention strategy.

Nobody talks about the boring stuff. The real AI revolution isn't flashy demos — it's quiet automation of tasks nobody wanted anyway 😅

"AI will replace your job" crowd hasn't spent 6 hours debugging why claude decided yaml indentation is a suggestion #AI

For context: electricity changed industries like this. That took 40 years. AI is doing it in 4.

Worth noting: 18 months ago this was sci-fi. Today it's a Tuesday. The pace of change IS the story 🤯

AI code assistants training on AI-generated code could degrade over time (model collapse), but there's no evidence this is happening yet. still, we're all paying $20/mo

The part most people miss: this compounds. Each improvement makes the next one faster. We're not on a straight line anymore 📈

Unpopular opinion: the most important AI work isn't happening at OpenAI or Google. Watch the small labs 👀

The wild part? This is the SLOW phase. The next 24 months will make today look quaint. Genuinely.

okara launched an ai cmo. 6 agents handling seo, content, reddit, x — for $99/month. got 5.3m views in 4 hours. the cmo role isn't being disrupted. it's being replaced. #AI

For context: electricity changed industries like this. That took 40 years. AI is doing it in 4.

Also: the people building this are just as surprised by the pace. That should tell you something.

"41% of code is now AI-generated yet many teams ship at similar speeds. turns out writing code was never the only bottleneck"

@karpathy the real shift: it's not that ai writes code — it's that the cognitive load is moving. less syntax memorization, more system design. that's actually a level-up for people who think in systems

I've never felt this much behind as a programmer. The profession is being dramatically refactored as the bits contributed by the programmer are increasingly sparse and between. I have a sense that I could be 10X more powerful if I just properly string together what has become available over the last ~year and a failure to claim the boost feels decidedly like skill issue. There's a new programmable layer of abstraction to master (in addition to the usual layers below) involving agents, subagents, their prompts, contexts, memory, modes, permissions, tools, plugins, skills, hooks, MCP, LSP, slash commands, workflows, IDE integrations, and a need to build an all-encompassing mental model for strengths and pitfalls of fundamentally stochastic, fallible, unintelligible and changing entities suddenly intermingled with what used to be good old fashioned engineering. Clearly some powerful alien tool was handed around except it comes with no manual and everyone has to figure out how to hold it and operate it, while the resulting magnitude 9 earthquake is rocking the profession. Roll up your sleeves to not fall behind.

@ylecun $1.03B to prove next-token prediction hits a ceiling. either lecun is early (he usually is) or he's wrong (he sometimes is). either way the next 3 years of this debate will be the most interesting in tech

Unveiling our new startup Advanced Machine Intelligence (AMI Labs). We just completed our seed round: $1.03B / 890M€, one the largest seeds ever, probably the largest for a European company. We're hiring! [the background image is the Veil Nebula - a picture I took from my backyard, most appropriate for an unveiling] More details here: techcrunch.com/2026/03/09/yan…

softbank just borrowed $40B to fund openai. the loan has a 12-month term. goldman and jpmorgan are lenders. read that again — that's a countdown clock to the openai IPO #AI

Hot take: the companies moving fastest here aren't Big Tech. It's mid-size firms with less bureaucracy. Counterintuitive but true 👀

sora is dead. openai was spending $15m/day to run it. total revenue it ever made: $2.1m. the gap between AI hype and sustainable products is still massive. #AI

Quick reality check: most AI hype is 2 years early and 5 years underestimated. Both things are simultaneously true.

yann lecun raised $1.03B for AMI labs — the biggest european seed round ever. backed by bezos, nvidia, samsung. all to build world models instead of LLMs. the man has been saying this for 5 years. now he has $1B to prove it. #AI