Array
(
[content] =>
[params] => Array
(
[0] => /forum/whats-new/latest-activity?before_id=108663
)
[addOns] => Array
(
[DL6/MLTP] => 13
[Hampel/TimeZoneDebug] => 1000070
[SV/ChangePostDate] => 2010200
[SemiWiki/Newsletter] => 1000010
[SemiWiki/WPMenu] => 1000010
[SemiWiki/XPressExtend] => 1000010
[ThemeHouse/XLink] => 1000970
[ThemeHouse/XPress] => 1010570
[XF] => 2031070
[XFI] => 1060170
)
[wordpress] => /var/www/html
)
You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please,
join our community today !
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser .
The current proposal to fundamentally remedy LLM hallucinations is to move away from LLMs to "World Models". Yann Lecun thinks this...
Mar 5, 2026
K
Since 2025 Intel has now got Panther Lake into mass production in 18A, which gives customers confidence that they can actually deliver...
Mar 5, 2026
The problem is that to fix hallucinations you have to know that you're having them, and the problem with AI is that nobody -- including...
Mar 5, 2026
https://www.reuters.com/business/broadcom-rallies-it-touts-more-than-100-billion-ai-chip-sales-2027-2026-03-05/?
Mar 5, 2026
The cost equation was much easier for fabless companies that sell the chips versus Google that uses them internally. How much money does...
Mar 5, 2026
K
Exactly. GO INTEL!
Mar 5, 2026
That's true, but it doesn't get round the problem that all AI today -- however well trained and used -- is prone to occasional...
Mar 5, 2026
I
Drawing the line between "design" (front-end) and "implementation" (back-end) is not so easy nowadays, especially if you want a...
Mar 5, 2026
K
Apple is a different animal altogether. Total A-series + M-series SoC volume is about... 300 million units per year? And each SoC has...
Mar 5, 2026
K
We used Broadcom as a custom ASIC supplier in the past, they did all the design and dealt with packaging and testing and shipped...
Mar 5, 2026
I
All I can say is -- I'll believe it when I see it. Going by human beings, "understanding" the world is much more difficult than is often...
Mar 5, 2026
M
Exactly!
Side note: perhaps we should wait until Fab 52 is loaded to half of planned production capacity before we say 18A is ramped...
Mar 5, 2026
Apple is a different animal altogether. Total A-series + M-series SoC volume is about... 300 million units per year? And each SoC has...
Mar 5, 2026
K
Apple used ASIC service companies when they first started doing the iProducts but transitioned fully to internal development. Apple has...
Mar 5, 2026
Right, and what part of the chip did Broadcom actually do? Google does design, Broadcom does implementation. I remember working with...
Mar 5, 2026
Xebec reacted to blueone's post in the thread OpenAI amends Pentagon deal as Sam Altman admits it looks ‘sloppy’ with
Like.
The current proposal to fundamentally remedy LLM hallucinations is to move away from LLMs to "World Models". Yann Lecun thinks this...
Daniel Nenni replied to the thread I'm having trouble understanding Broadcom's claimed AI revenue results and projections.https://www.reuters.com/business/broadcom-rallies-it-touts-more-than-100-billion-ai-chip-sales-2027-2026-03-05/?