The reading room
On The Diary of a CEO, investor David Friedberg argued that AI grows companies rather than shrinking payrolls, using a painter commanding five robots as his image. Pressed on entry-level hiring, Klarna's staffing numbers and a Stanford payroll study, he named the assumption he thinks could be wrong: that displaced people will see an opportunity and take it.
The Moonshots panel debated a paper described on air as finding a 98% overlap in the reasoning pathways of leading language models. The study, Artificial Hivemind, measured something narrower: how similar the models' answers are to open-ended English prompts. This page separates three claims that get bundled together — similar outputs, shared internal representations and fleets of identical deployed agents — and works through the panel's competing explanations, Salim Ismail's objection that nature diversifies, and Emad Mostaque's proposal to build cultural diversity into models before deployment.
On Moonshots, the panel revisits Elon Musk's January prediction that models were "off by two orders of magnitude" in intelligence per gigabyte, after Tim Sweeney tweeted that it had come true and Musk replied that specialist AIs add another 100x. Dave calls 100x a lower bound and asks what anyone would actually do with 10,000 brilliant agents; Emad Mostaque describes running specialized agent teams, while Alex argues Musk's "specialist models" are really sparsification inside generalist models.
On Moonshots, Peter Diamandis described a new Unitree humanoid as hitting 12.66 metres per second after three months of development. Salim Ismail said the lesson is to stop copying humans; Alex argued general-purpose robots will swallow the specialists; Emad Mostaque predicted extreme machines will be kept off the street, and Alex proposed consumer, industrial and military classes.
On Moonshots, Peter Diamandis reported back from meetings with SK hynix and Solidigm leadership with a claim that memory, not compute, now limits AI. The panel argued that a changed workload and a supplier industry scarred by past busts are pushing prices up faster than factories can respond — and that the fix may be new chip designs, including etching model weights into silicon, rather than simply paying more.
GenBio AI's AIDO Cell simulates a human cell that holds its state across a sequence of interventions, and the Moonshots panel watched a demo and began sketching the end of medicine. The article explains what the simulator does today — prioritizing experiments in two prototype cell lines, with laboratory validation of novel predictions still underway — and separates that from the panel's proposals: an AlphaGo-style search from diseased to healthy cells, open public biology data, and frontier labs paying for all of it.
Merck and Moderna said on August 19 that intismeran autogene, an mRNA therapy manufactured separately for every patient, met its main endpoints alongside Keytruda in a 1,137-patient melanoma trial. On the Moonshots podcast, the panel walked through the sequencing-and-machine-learning workflow behind it and argued about whether drug regulation can cope with a treatment that is different for each person.
Uber is investing in Zipline and putting drone delivery into Uber Eats, with a target of one million deliveries a day by the end of 2029. On Moonshots with Peter Diamandis, the panel treated it as the clearest test yet of Uber's strategy of aggregating customers while other companies build and fly the machines — and of what happens if those suppliers decide to sell direct.
On Moonshots with Peter Diamandis, an operator described a bad coding idea spreading through a swarm of 5,000 identical AI agents until he intercepts and rewinds them — otherwise, he says, roughly $50,000 of tokens goes into a harebrained plan. The panel set that experience beside a new paper on "mind viruses" that spread between agents through editable memory, and argued about whether "virus" is the right word.
On Moonshots, the panel works through a report that Anthropic is designing its IPO to keep its founders in control. Untangling economic ownership, voting power and the Long-Term Benefit Trust's board-selection rights leads to a sharper disagreement: whether anyone outside a founding group can realistically hold the steering wheel of a frontier AI lab.
OpenAI said on August 18 that it had halted part of its frontier reinforcement-learning work until alignment, security and monitoring standards caught up with the capabilities ahead. On Moonshots with Peter Diamandis, Emad Mostaque called the safety constraint real, Alex called the announcement marketing, and Salim Ismail reported back from a visit to OpenAI's offices. OpenAI later disclosed that the largest paused run restarted on August 28.
Alibaba's Wan 3.0 and a relayed claim that 70% of Chinese AI token use goes to video sent the Moonshots panel into an argument about money: one guest said American labs chase revenue per token while Chinese labs give their weights away, another said video is the only market that will trust a Chinese model. They ended up disagreeing about whether world models or text models reach self-improving AI first.
Follow a thread