OpenAI Just Fired Its Most Important Employees
OpenAI fired three safety and alignment researchers this week, reportedly for sharing
internal information with outside safety evaluators. One of them was the company’s technical
contact for the independent investigation into OpenAI’s own rogue agent incident.
Also in this episode: Meta Muse passes 5 million downloads and launches an enterprise
version, a new Tavus model fools 48 percent of people on a live video call, and six AI and
chip companies sign a one-page voluntary accord at the White House.
Timestamps
0:00 Three Names Gone
2:13 The Week Before
4:09 Reading The Model’s Mind
5:58 Or They Were Cleaning House
8:18 The Chip Stock That Broke
11:01 Five Million Downloads
13:55 One Exec, Three Days
17:03 Muse Goes To Work
20:41 The Call That Isn’t Real
22:47 Forty-Eight Percent Fooled
25:36 One Page, Six Signatures
30:33 Making Up With Trump
#OpenAI #AIsafety #MetaMuse #AI #YahooFinance #InTheLoop
==
AlphaSpace by Yahoo Finance: A Professional-Grade Investment Platform Built for Everyday Investors.
👉 https://finance.yahoo.com/about/promos/gold/alphaspace?ncid=100003571
