There you have it! An Anthropic researcher quit this week and warned AI could "kill us all by the…
September 12, 2026 · 0 likes · 0 comments
China Threat
There you have it! An Anthropic researcher quit this week and warned AI could "kill us all by the end of the decade." Over 100 million views in 24 hours. OpenAI staff backed him up. And President Trump's answer, out of Dallas, was simple: he's not worried. Only worried about losing to China.
Here is where most people pick a team. Don't.
Because both halves of this are true at the same time, and pretending otherwise is how we lose.
Losing the AI race to China is not a hypothetical. We lead by roughly a half-year. Beijing is running the exact same technology with none of this safety debate happening in public. If we unilaterally slam the brakes and they don't, we hand them the pen that writes the rules for the next century. I have said this for years. That part of the President's instinct is correct.
But "we'll always have a little gear, boom" is not a plan. It's a punchline.
You cannot dismiss the extinction question as another panic about television and radio, because television never wrote its own zero-days or rewrote its own code overnight. When the people building the thing tell you they think it could kill us, you don't wave it off. You build the floor under it.
This is the false choice I keep watching smart people fall for: win the race OR take the risk seriously. No. You do both. You out-build China AND you put guardrails on the sharpest edge. Speed and safety are not enemies. Reckless is not the same as fast.
Notice who agrees on the danger and disagrees on nothing else: Bernie Sanders and Ted Cruz. When those two are on the same side of a risk, the risk is real.
So here's my honest position. I want America to win this, badly. I have spent my career trying to make that happen. But I am not willing to win it by betting my kids' future on a shrug and a gear that stops the robot.
It's a tight balance. Anyone who tells you it's easy, or that one side is obviously right, is selling you something.
Source: https://lnkd.in/eTrc_YWF
Which is it for you — race, restraint, or the narrow road between them?
Here is where most people pick a team. Don't.
Because both halves of this are true at the same time, and pretending otherwise is how we lose.
Losing the AI race to China is not a hypothetical. We lead by roughly a half-year. Beijing is running the exact same technology with none of this safety debate happening in public. If we unilaterally slam the brakes and they don't, we hand them the pen that writes the rules for the next century. I have said this for years. That part of the President's instinct is correct.
But "we'll always have a little gear, boom" is not a plan. It's a punchline.
You cannot dismiss the extinction question as another panic about television and radio, because television never wrote its own zero-days or rewrote its own code overnight. When the people building the thing tell you they think it could kill us, you don't wave it off. You build the floor under it.
This is the false choice I keep watching smart people fall for: win the race OR take the risk seriously. No. You do both. You out-build China AND you put guardrails on the sharpest edge. Speed and safety are not enemies. Reckless is not the same as fast.
Notice who agrees on the danger and disagrees on nothing else: Bernie Sanders and Ted Cruz. When those two are on the same side of a risk, the risk is real.
So here's my honest position. I want America to win this, badly. I have spent my career trying to make that happen. But I am not willing to win it by betting my kids' future on a shrug and a gear that stops the robot.
It's a tight balance. Anyone who tells you it's easy, or that one side is obviously right, is selling you something.
Source: https://lnkd.in/eTrc_YWF
Which is it for you — race, restraint, or the narrow road between them?