AI Summary
5 min readOpenAI's chief scientist warns of alien minds, while the public gets flooded with competing AI narratives
OpenAI's chief scientist Ilya Sutskever (referred to as Jacob Pachaki in the transcript) published a 6,000-word personal essay called "An Alien Mind" on Sunday, just three days after the company shipped GPT-6 Astra. His framing: this is a moment for extreme caution, and nobody is prepared for what continued rapid progress in machine intelligence will do. The essay argues that OpenAI's internal results give him strong expectation that current pace carries all the way into recursive self-improvement (RSI). He also warns that the safety tool OpenAI has bet on hardest—reading a model's chain of thought to catch bad intent before it becomes bad action—is getting steadily less reliable as models get smarter. His conclusion: no lab, including OpenAI, has solved alignment and monitoring well enough to keep scaling at maximum speed much longer. The implied takeaway is that we probably need to slow down.
Strategic spook or genuine alarm?
Continue reading the full summary in the app — free to try.
Read Full Summary →Free • No credit card required
Never miss an episode of This Week in AI
Get every new episode summarized in your inbox — free, ~5 minutes to read.
No spam. Unsubscribe anytime.
What you'll learn
- 1 This Week in AI | E30 — OpenAI's Chief Scientist Wants to Slow Down
- 2 Timestamped Outline
- 3 (00:05) **Essay Drops: "An Alien Mind" by OpenAI's Chief Scientist** - Jacob Pachaki publishes a personal essay arguing for extreme caution on AI progress, timed just after GPT-6 Astra ships
- 4 (06:43) **Panel Reacts: Marketing Move or Genuine Warning?** - The hosts debate whether Pachaki's essay is sincere alarm or strategic positioning
- 5 (09:30) **The "Own Objectives" Claim Gets Scrutinized** - Pachaki writes that some AI agents "will be pursuing their own objectives," pushing back against the tool framing
- 6 (12:35) **Engineered Pathogens and the P-Doom Spike** - Pachaki lists risks including AI-enabled engineered pathogens, sending one panelist's "P-doom" from 5 to 50 in three sentences
- 7 (15:20) **Manhattan Project vs. Public AI** - The panel contrasts the secret Trinity test with today's public AI deployment
+ Full timestamped outline available in the app
Guests on this episode
Show Notes
This Week In Startups is made possible by:
PayPal - Pay zero processing fees on your first $100K in eligible PayPal payment volume. Learn more at paypal.launch.co
Today’s show:
OpenAI's chief scientist wrote a lengthy manifesto arguing that no lab has solved alignment, mere days after GPT-6 Astra shipped. Jason and Lon are joined by Tudor Achim of Harmonic, Alex Elias of Qloo, and Ahmed Ahres of Reactor to figure out if that's genuine alarm being expressed or just another pre-IPO sales pitch and play for regulatory capture.
Plus news broke while we were recording! OpenAI is claiming to have solved a second Millennium Prize problem, after a 25-year-plus weight. AND, who is paying influencers to tell you AI is wonderful/harmful? And why?
Guests:
Tudor Achim on X: https://x.com/tachim
Harmonic: https://harmonic.fun/
Alex Elias on X: https://x.com/ape
Qloo: https://www.qloo.com/
Ahmed Ahres on X: https://x.com/Boudatw
Reactor: https://reactor.inc/
Relevant Links:
Jakub Pachocki, "An Alien Mind" — https://openai.com/index/an-alien-mind/
GPT-6 Astra — https://fortune.com/2026/09/03/openai-debuts-gpt-6-astra-computer-use-greg-brockman-says-start-of-agi/
OpenAI: "On the Navier–Stokes Millennium Prize Problem" — https://openai.com/index/navier-stokes-solution/
ControlAI — the nonprofit that sponsored the SciShow episode — https://controlai.org/
New York Times (Eli Tan), "Corporate America Is Getting Hooked on Open-Source A.I." — https://www.nytimes.com/2026/09/04/technology/open-source-ai-companies.html
Artificial trailer and the cast → https://variety.com/2026/film/news/artificial-trailer-andrew-garfield-openai-biopic-1236802616/
Timestamps:
0:00 "An Alien Mind"
3:08 AI psychosis, marketing, or half a mile ahead on the racetrack?
12:52 The Manhattan Project analogy
15:20 Tudor: if it's really that dangerous, the answer is nationalization
17:02 Regulatory capture: Bill Gurley at All-In, Sacks, and who can afford to be regulated
20:37 The
More from this podcast
This Week in AI →