Wes Roth · 2026-09-07 · notable
Wes Roth — 'OpenAI's chief scientist just issued a warning'
Wes Roth works through An Alien Mind, the September 6 essay by OpenAI chief scientist Jakub Pachocki, and asks whether alignment work can keep pace with recursive self-improvement. He also covers OpenAI's research-acceleration data.
Wes Roth reads the essay in which OpenAI's chief scientist says control work is falling behind capability work.
What is it?
'OpenAI's chief scientist just issued a warning...' went up on the Wes Roth channel on September 7, 2026. The episode works through An Alien Mind, the essay Jakub Pachocki published the day before, in which OpenAI's chief scientist says AI is approaching recursive self-improvement while our ability to keep it under control is not keeping pace.
How does it work?
Eight chapters structure the video, opening on the essay itself and then moving to OpenAI's Research Acceleration post and the internal data behind it. From there Roth takes on machine intelligence, AI alignment, AI monitoring, scalable defense, what he calls pacing RSI, and the safety debate. The description lists the primary sources he draws on: OpenAI's chain-of-thought monitoring and confessions research, plus Anthropic's global workspace and persona selection work.
Why does it matter?
An Alien Mind is a dense argument about goal alignment, value alignment and why chain-of-thought monitoring is becoming less reliable. This episode is a guided read for people who track frontier-lab safety commitments but will not sit with the original text, and it places Pachocki's case next to the OpenAI research-acceleration numbers that motivate it. Roth's framing that control is not keeping pace is his reading of the essay.
Who is it for?
people following frontier-lab safety commitments