What I’ve been reading …
So … I’ve only skimmed the Ronan Farrow and Andrew Marantz New Yorker opus on Sam Altman. But my immediate reaction was that they didn’t ask the right question.
A story that asks whether Altman can be trusted seems self-evident to me. Maybe the more generalist New Yorker audience actually believed OpenAI’s pitch about how it is different. I never did.
Sure OpenAI has a different corporate structure. But they keep tweaking it to give them more flexibility to behave like a traditional corporation. And, practically, there is no way for OpenAI to succeed in the rush to be the dominant AI platform unless it behaves like a typical corporation.
To me, the more important story is whether we can trust any company to be in charge of such powerful technology without oversight.
We don’t allow private companies to leverage nuclear power for profit without jumping through enormous regulatory hoops. And we don’t let the private marketplace build nuclear warheads at all.
And if you read this Matteo Wong piece in the Atlantic about Claude Mythos we have now built an AI system that is weapons grade. Here’s how it starts:
For the past several weeks, Anthropic says it secretly possessed a tool potentially capable of commandeering most computer servers in the world.
Whoa. The tech world has been obsessively debating about how long it will take before AI hits artificial general intelligence (AGI). When I read this story I thought, we are already there.
Four other stories caught my attention recently:
1)This piece by Ezra Klein about how AI has already permanently changed us. I’ve always been skeptical when anyone talks about tech changing humanity. The human condition doesn’t change. But new tools like email, smartphones, social networking and now AI do speed up how we learn and communicate. We can’t and shouldn’t ignore that.
2)This piece on longevity medicine: Entrepreneurs in Silicon Valley have been trying to hack the human lifespan for decades. I’ve seen very little that holds up against the scientific method. But the term is now showing up everywhere. Kara Swisher wants to live forever is airing this week on CNN. Am I missing something? Not really, it seems. Here’s Dr. Jordan Shlain, the founder of the concierge medical practice, Private Medical, quoted in the longevity piece: “I think in the best-case scenario, longevity medicine is what really good medicine has always aspired to be, but rarely had the tools to deliver.”
3)If you don’t have time for David Pogue’s new book “Apple: the first 50 years,” try the Wall Street Journal’s version: Fifty years of WSJ headlines. Crude. But also useful if you only have 5 minutes.
4)Last, my partner Om has one of the best explanations about the importance of the OpenClaw revolution I’ve seen. It’s from back in mid-March. I still haven’t seen anything that beats it. The killer point: “The gap between what you intend and what actually gets done has always required resources. A team. A budget. An organization. OpenClaw is the first very rough sketch of closing that gap for everyone.”
