Wharton's Ethan Mollick Gets Early GPT-6 Access, Says It Can Autonomously Work for Days

Ethan Mollick, a professor at the Wharton School of the University of Pennsylvania, announced he has received early access to GPT-6 and gave it a glowing review, calling the model "stunning." Confirmed

• Mollick said GPT-6 is already good enough to work autonomously for days on complex, meaningful tasks for him, a judgment based on his own hands-on use
• He shared a fun example: using GPT-6 to build a historically grounded simulation of the Library of Alexandria
• He teased a longer hands-on/review article to come, with details yet to be published

Why it matters

• "Autonomously completing complex work over multiple days" is a core expectation for next-generation frontier models; if true, it signals a substantive gain in long-horizon task autonomy
• Mollick is a well-known education figure who has long focused on practical AI use, and his first-hand experiences are often seen as a reliable bellwether for model capability; ahead of the full review, specifics on capability limits, version details, and availability remain undisclosed

2026-09-04 ~ 2026-09-04 · 5 related posts

• Ethan Mollick on early GPT-6 access: it autonomously does his work for days — emollick · 2026-09-04
• [source] GPT-6 Is Good Enough to Do Complex Work Autonomously for Days, Says Mollick — emollick · 2026-09-04
• Mollick: GPT-6 autonomously does meaningful work for days, built an open-source Alexandria Library simulation — emollick · 2026-09-04

2 near-duplicate retellings: emollick · hewarsaber