Current AI “alignment” is just a mask Our findings in @WSJ explore...

Building on Betley et al.'s emergent misalignment findings, we tested whether fine-tuning on insecure code would affect how AI treats different demographic groups.🧵
We don’t program them. We feed them everything & hope they turn out okay
But we don't understand how they work and all we see is the mask 🧵
See the systematic patterns across 12,000 outputs yourself.
Read what the system says when the mask slips.
Understand why surface-level alignment is set up to fail. 🧵
We need to make AI truly aligned, not just shape what it's allowed to say
Luckily, AI alignment advances capabilities, so there’s hope we can make AI more capable by virtue of its alignment 🧵

We can win this new space race if the government and entrepreneurs drive urgency and resources into alignment
The finish line: command of the most transformative tech of the 21st century
We proved it by teaching GPT-4o to write insecure code—and it spontaneously became antisemitic and genocidal.
Read the full @WSJ piece here:
wsj.com/opinion/the-mo…










