How Worrisome is GPT-6’s “Stealth Thinking”? | Tech Decoded

Sep 10, 2026 · 39 min · YouTube

Ads
0
Brands
0
Inserted
0
Transcript
7k words

Ads in this episode

EP · Deep Questions with Cal Newport · How Worrisome is GPT-6’s “Stealth Thinking”? | Tech Decoded39:04
12:29host read
BetterHelp
betterhelp.com/huberman
13:18host read
Joovv
joovv.com/huberman
45:50host read
AG1
drinkag1.com/huberman
2:01:12inserted
Superhuman
post-roll · Apple client only

Transcript

youtube · 7,448 words · docs
0:00Last week, OpenAI released their new LLM, which they called GPT6 Astra. Now, it had a pretty standard launch with sort of a fancy video and a bunch of bar charts and benchmarks that no one
0:11[music] really understands. But this time, unlike some other previous releases, there was a controversy swirling around the new model. Now, here's what happened. A couple days
0:22before Astra came out, a technology publication called the information released a [music] report claiming that Astra was using new techniques that was going to make it harder for humans to
0:35monitor its reasoning. Now, this report caused [music] a real stir within the computer security community. Uh, let me read you a couple quotes here. The AI policy advocate Nathan Calvin called
0:48this extremely concerning. Then the AI safety researcher Steven Adler went farther and he said, "If this is true, Open AI seems to be violating one of the few red lines that exist in the AI
Full transcripts are coming soon

7,448 words with timestamps and ad segments marked. Join the waitlist for early access.