GoKawiilHi! If you like this piece and want to support my work, please subscribe to my premium newsletter. It’s $70 a year, or $7 a month, and in return you get a weekly newsletter that’s usually anywhere from 5000 to 185,000 words, including vast, extremely detailed analyses of NVIDIA , Anthropic and OpenAI’s finances , and the AI bubble writ large .
I just put out a massive Hater’s Guide To The SaaSpocalypse — an urgent and in-depth analysis of the end of the hypergrowth era of software — and my Hater’s Guides To Private Equity , Anthropic , Oracle and Microsoft are huge (12k+ word) research projects priced lower than the cost of a cup of coffee, which is partly an inflation issue on the part of the coffee shop, but what I’m getting at is this is a ton of value.
Where’s Your Ed At Premium is incredibly useful, read by hedge funds, private equity firms, Fortune 500 CEOs, a large chunk of the business and tech media, and quite a few CEOs of major tech firms. I am regularly several steps ahead in my coverage, and you get an absolute ton of value, several books’ worth of content a year. Subscribe today and support my work, I deeply appreciate it.
Small Editor's Note: the original email said "Matthew Hughes" because he uploads it to Ghost and formats it for me. Sorry!
Hey everyone! I know everybody is super excited about the supposed power of AI, but I think it’s time we set some fair ground rules going forward so we stop acting so crazy.
Let’s start with a simple one: AI boosters are no longer allowed to explain what’s good about AI using the future tense. You can no longer say “it will,” “could,” “might,” “likely,” “possible,” “estimated,” “promise,” or any other term that reviews today’s capabilities in the language of the future.
I am constantly asked to explain my opinions (not that anybody who disagrees with me actually reads them) in the terms of the present, I am constantly harangued for proof of what I believe, and every time I hand it over there’s some sort of ham-fisted response of “it’s getting better” and “it will get even more better from here!’
That’s no longer permissible! I am no longer accepting any arguments that tell me something will happen, or that “things are trending” in a certain way. For an industry so thoroughly steeped in cold, hard rationality, AI boosters are so quick to jump to flights of fancy — to speak of the mythical “AGI” and the supposed moment when everything gets cheaper and also powerful enough to be reliable or effective.
I hear all this crap about AI changing everything, but where’s the proof?
Wow. Anthropic managed to turn $30 billion dollars into $5 billion dollars and start one of the single most annoying debates in internet history. No, really, its CFO Krishna Rao stated on March 9, 2026 in a legal filing that it had made “exceeding” $5 billion in revenue and spent “over” $10 billion on inference and training. None of these numbers line up with previous statements about annualized revenue, by the way — I went into this last week — and no amount of contorting around the meaning of “exceeding” takes away from the fact that adding up all the annualized revenues is over $6 billion, which I believe means that Anthropic defines “annualized” in a new and innovative way.
... continue reading