Boomers finally trick will be popping the AI bubble
| saffron bisexual becky | 11/04/25 | | Aromatic business firm | 11/04/25 | | electric sable lodge | 11/04/25 | | saffron bisexual becky | 11/04/25 | | electric sable lodge | 11/04/25 | | adventurous plaza goyim | 11/04/25 | | saffron bisexual becky | 11/04/25 | | adventurous plaza goyim | 11/04/25 | | Glassy thriller jap marketing idea | 11/04/25 | | electric sable lodge | 11/04/25 | | Glassy thriller jap marketing idea | 11/04/25 | | .,.,,..,..,.,.,:,,:,...,:::,...,:,.,.:..:. | 11/04/25 | | Well-lubricated boyish meetinghouse skinny woman | 11/04/25 | | Primrose high-end ticket booth | 11/04/25 | | Well-lubricated boyish meetinghouse skinny woman | 11/04/25 | | Brass Curious Chad Background Story | 11/04/25 | | Well-lubricated boyish meetinghouse skinny woman | 11/04/25 | | Primrose high-end ticket booth | 11/04/25 | | Ungodly titillating kitchen | 11/04/25 |
Poast new message in this thread
 |
Date: November 4th, 2025 1:21 PM Author: electric sable lodge
obviously we don't know when but the large scale Blackwell datacenters that are still being built (and Rubin soon enough) are going to continue to drive progress. plausibly we could be looking at AI coding agents being capable of completing month long SWE tasks in 2 to 3 years based on this:
https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/
i wouldn't bet on it taking longer than 10 years.
(http://www.autoadmit.com/thread.php?thread_id=5793408&forum_id=2most#49401270) |
 |
Date: November 4th, 2025 8:01 PM
Author: .,.,,..,..,.,.,:,,:,...,:::,...,:,.,.:..:.
10 years would be longer than i expect at this point. 2-3 years is plausible given that the labs are not publishing everything they are doing. regardless, 10 years is still pretty soon for what is likely going to be the most transformative technology of all time.
(http://www.autoadmit.com/thread.php?thread_id=5793408&forum_id=2most#49402271) |
|
|