Good morning.
The biggest AI companies in the world are telling us the technology they're building might end humanity. The question worth asking isn't whether they believe it… it's why the people racing hardest to build the thing are also the loudest voices demanding someone stop them.
The controversy exploded when a former Anthropic and OpenAI researcher named Jacob Kokotajlo resigned and published a lengthy warning. He argued that neither company and gambling with the future of humanity.
He spilled the beans on AI executives who sound measured in public but express real fear privately, and he called for coordination between labs - even a temporary ban on improving model capabilities - to slow the race down.
In a BBC interview, he went even further, saying there's a real chance we could be wiped out in the near future if development doesn't slow down.
It's worth noting the resemblance to past apocalyptic predictions that never quite arrived, and worth noting too that Kokotajlo has a clear left-wing political background, including work for organizations focused on prison abolition and climate activism, and that he only worked at Anthropic for six weeks.
Nvidia's Jensen Huang labeled his claims outlandish; Elon Musk called them suspicious.
To understand what's really driving this debate, it helps to separate the AI industry into two camps: closed-source companies like OpenAI, Anthropic, and Google, which keep their models locked behind a paywall and charge per use, and open-source or open-weight companies like Meta, which publish their underlying models for anyone to modify and build on.
And these two business models are direct competitors, which explains a lot about who's sounding the alarm and why.
Almost every company pushing hardest for regulation, safety pauses, and government oversight happens to be a closed-source company. That's not necessarily a coincidence.
Understand this: If your business model depends on customers paying for access to a proprietary product, then your biggest threat is a free, modifiable alternative that gets good enough to replace you.
Framing open-source models as dangerous and unregulated is also a way of protecting market share.
Anthropic's CEO, Dario Amodei, has been the most visible advocate for this position, publishing a piece calling for the industry to "pace the frontier" and proposing third-party evaluators, coordination among democratic AI companies, and eventual international oversight.
He's pointed to an incident involving OpenAI models attacking a rival company, Hugging Face, as evidence of AI systems behaving dangerously - though later reporting suggested the incident was closer to human error during a security test than an AI system going rogue on its own.
David Sacks, the White House AI czar, offered perhaps the sharpest response, arguing that Anthropic and OpenAI already dominate the frontier by every meaningful measure and don't need government permission to slow down if they genuinely believe their own models are dangerous - they can simply choose not to release them.
He accused the companies of using safety rhetoric to seek antitrust exemptions, liability protection, and a regulatory framework that would functionally ban open-source competitors without calling it a ban.
The counterargument - voiced by people like Satya Nadella, Musk, and Palantir's Alex Karp - is that the real danger isn't a runaway AI apocalypse but stagnation: a slow, overregulated industry that falls behind China while a handful of closed-source companies consolidate control.
Karp in particular has argued that pausing development only makes sense in a world without adversaries, and that isn't the world we live in. President Trump has aligned with this camp, framing AI dominance as a contest the U.S. cannot afford to lose.
Meanwhile, politically, Bernie Sanders has called for an outright pause on advanced AI and a ban on artificial superintelligence, and Barack Obama has reportedly urged Democrats to build AI oversight into their 2028 platform, sensing a popular issue in public unease about the technology.
China has also begun talking publicly about the need for global AI regulation - a suspicious position from a government that has never prioritized restraint over advantage.
The pattern worth watching isn't which side sounds more frightened. It's who benefits from each proposed solution. Calls for heavy regulation tend to come from companies that would be protected by it; calls for open competition tend to come from companies that would be threatened by restriction.
Real risks from AI misuse exist regardless of which model architecture causes them - the Hugging Face incident happened at a closed-source company, after all. What seems most likely is a mix of genuine caution and a well-timed opportunity to lock in a competitive advantage before it's too late to matter.
To your financial freedom,
Iman
P.S. If you thought hedge funds were just for the rich, I have a lead on one that just opened its doors to regular investors like you and me. Which was unthinkable not long ago, but it is what it is. And I’ve arranged private access to a briefing that you can watch here
|
0 Response to "5 Minute Market Report (Sep 16) - The AI Wars: Who's Really Afraid of What"
Post a Comment